Vous êtes sur la page 1sur 450

WELDING PROCESSES

Edited by Radovan Kovacevic









Welding Processes
http://dx.doi.org/10.5772/2884
Edited by Radovan Kovacevic

Contributors
J. Zhou, H.L. Tsai, Hitoshi Ozaki, Muneharu Kutsuna, Kelvii Wei Guo, Ikhwan Naim Md Nawi,
Jalil Ali, Mohamed Fadhali, Preecha P. Yupapin, Abdel-Monem El-Batahgy, Junjie Ma, Fanrong
Kong, Blair Carlson, Radovan Kovacevic, Zhiguo Gao, Jauhari Tahir Khairuddin, Jamaluddin
Abdullah, Zuhailawati Hussain, Indra Putra Almanar, Farag Soul, Nada Hamdy, Miroslav
Mijajlovi, Dragan Mili, Yoji Ogawa, Marek Stanisaw Wglowski, Wei Jie Zhang, Yu Kang
Liu, Yu Ming Zhang, Riyadh Mohammed Ali Hamza, P. S. Wei, Clbio Domingues da Silveira-
Jnior, Morgana Guilherme de Castro, Letcia Resende Davi, Flvio Domingues das Neves,
Verediana Resende Novais, Paulo Czar Simamoto-Jnior

Published by InTech
Janeza Trdine 9, 51000 Rijeka, Croatia

Copyright 2012 InTech
All chapters are Open Access distributed under the Creative Commons Attribution 3.0 license,
which allows users to download, copy and build upon published articles even for commercial
purposes, as long as the author and publisher are properly credited, which ensures maximum
dissemination and a wider impact of our publications. After this work has been published by
InTech, authors have the right to republish it, in whole or part, in any publication of which they
are the author, and to make other personal use of the work. Any republication, referencing or
personal use of the work must explicitly identify the original source.

Notice
Statements and opinions expressed in the chapters are these of the individual contributors and
not necessarily those of the editors or publisher. No responsibility is accepted for the accuracy
of information contained in the published chapters. The publisher assumes no responsibility for
any damage or injury to persons or property arising out of the use of any materials,
instructions, methods or ideas contained in the book.

Publishing Process Manager Viktorija Zgela
Typesetting InTech Prepress, Novi Sad
Cover InTech Design Team

First published November, 2012
Printed in Croatia

A free online edition of this book is available at www.intechopen.com
Additional hard copies can be obtained from orders@intechopen.com


Welding Processes, Edited by Radovan Kovacevic
p. cm.
ISBN 978-953-51-0854-2








Contents

Preface IX
Section 1 Laser Welding 1
Chapter 1 Hybrid Laser-Arc Welding 3
J. Zhou and H.L. Tsai
Chapter 2 Dissimilar Metal Joining of Zinc Coated Steel
and Aluminum Alloy by Laser Roll Welding 33
Hitoshi Ozaki and Muneharu Kutsuna
Chapter 3 In situ Reaction During Pulsed Nd:YAG Laser Welding
SiC
p
/A356 with Ti as Filler Metal 55
Kelvii Wei Guo
Chapter 4 Nd:YAG Laser Welding
for Photonics Devices Packaging 75
Ikhwan Naim Md Nawi, Jalil Ali,
Mohamed Fadhali and Preecha P. Yupapin
Chapter 5 Laser Beam Welding of Austenitic Stainless
Steels Similar Butt and Dissimilar Lap Joints 93
Abdel-Monem El-Batahgy
Chapter 6 Mitigating Zinc Vapor Induced Weld Defects in Laser
Welding of Galvanized High-Strength Steel by Using
Different Supplementary Means 117
Junjie Ma, Fanrong Kong, Blair Carlson and Radovan Kovacevic
Section 2 Numerical Modeling of Welding Processes 139
Chapter 7 Numerical Modelling to Understand
Cracking Phenomena During Laser-GMA
Hybrid Welding Nickel-Base Superalloys 141
Zhiguo Gao
VI Contents

Chapter 8 Development of a Comprehensive Process
Model for Hybrid Laser-Arc Welding 165
Fanrong Kong and Radovan Kovacevic
Chapter 9 Principles and Thermo-Mechanical
Model of Friction Stir Welding 191
Jauhari Tahir Khairuddin, Jamaluddin Abdullah,
Zuhailawati Hussain and Indra Putra Almanar
Chapter 10 Numerical Simulation of Residual Stress
and Strain Behavior After Temperature Modification 217
Farag Soul and Nada Hamdy
Chapter 11 Analytical Model for Estimating the Amount of Heat
Generated During Friction Stir Welding: Application
on Plates Made of Aluminium Alloy 2024 T351 247
Miroslav Mijajlovi and Dragan Mili
Section 3 Sensing of Welding Processes 275
Chapter 12 Visual Analysis of Welding Processes 277
Yoji Ogawa
Chapter 13 Monitoring of Arc Welding Process
Based on Arc Light Emission 305
Marek Stanisaw Wglowski
Chapter 14 Real-Time Measurement of Three Dimensional
Weld Pool Surface in GTAW 333
Wei Jie Zhang, Yu Kang Liu and Yu Ming Zhang
Section 4 General Topics in Welding 367
Chapter 15 Optimized Stud Arc Welding Process Control Factors
by Taguchi Experimental Design Technique 369
Riyadh Mohammed Ali Hamza
Chapter 16 The Physics of Weld Bead Defects 395
P. S. Wei
Chapter 17 Welding Techniques in Dentistry 415
Clbio Domingues da Silveira-Jnior, Morgana Guilherme
de Castro, Letcia Resende Davi, Flvio Domingues das Neves,
Verediana Resende Novais and Paulo Czar Simamoto-Jnior








Preface

Despite the wide availability of literature on welding processes, a need exists to
regularly update the engineering community on advancements in joining techniques
of similar and dissimilar materials, in their numerical modeling, as well as in their
sensing and control. In response to InTechs request to provide undergraduate and
graduate students, welding engineers, and researchers with updates on recent
achievements in welding, a group of 34 authors and co-authors from 14 countries
representing five continents have joined to co-author this book on welding processes,
free of charge to the reader.
This book is divided into four sections: Laser Welding; Numerical Modeling of
Welding Processes; Sensing of Welding Processes; and General Topics in Welding.
The first section, Laser Welding, includes six chapters covering topics ranging from an
extensive overview of the physics of hybrid laser-arc welding, to welding of dissimilar
materials, aluminum matrix composite, photonics devices, stainless steel, and high-
strength galvanized steels.
The second section, Numerical Modeling of Welding Processes, includes five chapters
focused on numerical modeling of hybrid laser-arc welding, friction stir welding, and
modeling of residual stresses and distortions with an overview of techniques designed
to mitigate residual stresses during welding.
The third section, Sensing of Welding Processes, includes three chapters focused on
applying high-speed imaging to the study of fusion welding processes, using arc light
emission to monitor the welding process, and measuring in real time the free surface
of a molten pool in GTAW (gas tungsten arc welding).
The final section, General Topics in Welding, includes three chapters focused on
optimization of stud arc welding process parameters, weld bead defects, and welding
techniques in dentistry.
It is our firm belief that the material presented in this book will appeal to a broad
audience of readers interested in recent advancements in hybrid laser-arc welding,
X Preface

autogenous laser welding, friction stir welding, numerical modeling of different
welding processes, and real-time sensing of the welding processes.

Radovan Kovacevic, Ph.D., FAWS, FSME, FASME
Herman Brown Chair in Engineering
Professor of Mechanical Engineering
Director of the Research Center for Advanced Manufacturing
and Center for Laser-aided Manufacturing
Southern Methodist University, Dallas, TX,
USA




Section 1




Laser Welding



Chapter 1




2012 Zhou and Tsai, licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Hybrid Laser-Arc Welding
J. Zhou and H.L. Tsai
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/50113
1. Introduction
Hybrid laser-arc welding has received increasing interest in both academia and industry in
last decade
1,2
. As shown in Fig. 1, hybrid laser-arc welding is formed by combining laser
beam welding and arc welding. Due to the synergic action of laser beam and welding arc,
hybrid welding offers many advantages over laser welding and arc welding alone
3-6
, such as
high welding speed, deep penetration
7
, improved weld quality with reduced susceptibility
to pores and cracks
8-16
, excellent gap bridging ability
17-22
, as well as good process stability
and efficiency, as shown in Fig. 2.

Figure 1. Schematic sketch of a hybrid laser-arc welding process.

Welding Processes 4

Figure 2. Comparison between (a) a laser weld and (b) a hybrid laser-arc weld in 250 grade mild
steel.
Development of the hybrid laser-arc welding technique can be divided into three stages
1
.
The concept of hybrid laser welding was first proposed by Steen et al.
3, 23, 24
in the late
seventies. In their studies, a CO2 laser was combined with a tungsten inert gas (TIG) arc for
welding and cutting applications. Their tests showed clear benefits of combining an arc and
a laser beam in the welding process, such as a stabilized arc behavior under the influence of
laser radiation; a dramatic increase in the speed of welding of thin metal sheets; and an
increase in penetration depth compared with laser welding. Japanese researchers continued
Steens effort and developed various methods and corresponding devices for laser-arc
welding, cutting, and surface treatment. However, these efforts did not advance this joining
technique into engineering applications particularly because laser welding itself was not yet
an economic and viable joining technique at that time
25
. In the second stage of the
development of the hybrid laser-arc welding technique, the observed influence of the arc
column behavior by laser radiation was used to improve the efficiency of arc welding
processes, which leads to the laser-enhanced arc welding technology
1
. A characteristic
feature of this technology was that only a low-intensity laser beam was needed, i.e., the
required laser power was small compared to the arc power. For TIG welding, Cui and
Decker
26-28
demonstrated that a low-energy CO2 laser beam with a power of merely 100 W
could facilitate arc ignition; enhance arc stability; improve weld quality; and increase
welding speed due to a reduced arc size and higher arc amperages. However, despite such
reported improvements of the arc welding process through laser support, there were neither
subsequent extensive investigations of this subject nor known industrial applications of the
laser-enhanced arc welding technology. The third stage of hybrid welding technology
started in the early 1990s with the development of combined welding processes using a
high-power laser beam as the primary and an additional electric arc as the secondary
heating source
29-37
. At that time, although the continuous wave CO2 laser welding process
was already well established in industry, it had some known disadvantages, e.g., high
requirements of edge preparation and clamping; fast solidification leading to material-
dependent pores and cracks; as well as the high investment and operating costs for the laser
equipment. Additionally, some welding applications of highly practical interest could not be

Hybrid Laser-Arc Welding 5
solved satisfactorily by the laser welding process alone, e.g., joining of tailored blanks in the
automotive engineering; welding of heavy plates in shipbuilding industry; as well as high
speed welding of crack-susceptible materials. In searching for suitable solutions, the hybrid
laser welding was developed into a viable joining technique with significant industrial
acceptance during the last decade.
According to the combination of various heating sources used, hybrid welding can be
generally categorized as: (1) laser-gas tungsten arc (GTA) welding; (2) laser-gas metal arc
(GMA) welding; and (3) laser-plasma welding
25
. Since laser welding offers deep
penetration, primary heating sources commonly used in hybrid welding are CO2,
Nd:YAG, and fiber lasers. The first two types of lasers are well established in practice and
used for various hybrid welding process developments. While the fiber laser is still in
development for industrial applications, it seems to be a future primary heating source for
hybrid welding due to its high beam quality. The secondary heating sources used in
hybrid welding are mainly electric arcs. Dedicated processes can be divided into GMA
welding with consumable electrodes and GTA welding with non-consumable tungsten
electrodes. In GMA welding, the arc is burning between a mechanically supplied wire
electrode and the workpiece. The shielding gas used in GMA welding was found to have
significant effects on arc shape and metal transfer
38,39
. Hence, GMA welding can be
subdivided into metal inert-gas (MIG) and metal active-gas (MAG) welding according to
the type of shielding gas used. In GTA welding, a chemically inert gas, such as argon or
helium, is often used. A special form of this is the plasma arc welding (PAW), which
produces a squeezed arc due to a special torch design and results in a more concentrated
arc spot.
In hybrid welding, laser and arc are arranged preferably in a way that they can compensate
and benefit from each other during the welding process, which implies the creation of a
common interaction zone with changed characteristics in comparison to the laser welding
and the arc welding alone. In contrast to this is the arrangement in which laser and arc are
serving as two separate heating sources during the welding process. Several configurations
have been proposed. In a parallel arrangement, there is a distance in either the vertical or
horizontal direction along the path between both heating sources. In a serial arrangement,
the primary and secondary heating sources are moved along the same welding path with a
certain working distance, and the secondary heating source can either lead or follow the
primary heating source
1
. The first one enables a preheating of the region to be welded. It can
increase the efficiency of the laser welding process because materials to be welded are
locally preheated and energy losses through heat conduction are reduced. In comparison,
the second one often acts like a short-time post-heat treatment of the weld that can change
the weld microstructure favorably. There exists a key difference between parallel and serial
process arrangement. In a serial arrangement, additional energy is dissipated within the
weld seam region, whereas in the parallel arrangement, the heat flow is reduced only across
the weld seam. The option to move the working area temporally enables flexibility in
influencing the cooling rates in order to avoid defects.

Welding Processes 6
In the hybrid laser-arc welding process, the workpiece is first heated up and melted due
to the laser irradiation. The plasma arc between the consumable electrode and the
workpiece continues to heat up and melt the base metal and the droplets generated at the
electrode tip periodically detach and impinge onto the workpiece. Then a cavity with
large depth-to-width ratio called keyhole was formed in the weld pool under the
dynamical interaction of laser irradiation, plasma arc and filler droplets. An externally
supplied shielding gas provides the protection of molten metal from exposing to the
atmosphere. The successive weld pools create a weld bead and become a part of a welded
joint when solidified. The numbers of process parameters are greatly increased in the
hybrid welding, mainly including laser beam parameters, electric power parameters,
laser-arc interval, electrode diameter, wire feed speed, welding speed and shielding gas.
Bagger and Olsen
66
reviewed the fundamental phenomena occurring in laser-arc hybrid
welding and the principles for choosing the process parameters. Ribic et al.
67
reviewed the
recent advances in hybrid welding with emphases on the physical interactions between
laser and arc, and the effects of the combined laser-arc heat source on the welding
process.
Current understanding of hybrid laser-arc welding is primarily based on experimental
observations. Hybrid laser-arc welding is restricted to specific applications, predominantly
the joining of thick section plain carbon steels. In order to expand the applications of this
joining technique and optimize the processes for its current applications, fundamental
understanding of the transport phenomena and the role of each parameter becomes critical.
Numerical investigations were often carried out for this purpose. Ribic et al.
67
developed a
three-dimensional heat transfer and fluid flow model for laser-GTA hybrid welding to
understand the temperature field, cooling rates and mixing in the weld pool. Kong and
Kovacevic
68
developed a three-dimensional model to simulate the temperature field and
thermally induced stress field in the workpiece during the hybrid laser-GTA process.
Mathematical models have also been developed to simulate the weld pool formation and
flow patterns in hybrid laser-GMA welding by incorporating free surfaces based on the VOF
method. Generally, the typical phenomena in GMA welding such as droplets impingement
into the weld pool, electromagnetic force in the weld pool and the typical phenomena in
laser beaming welding such as keyhole dynamics, inverse Bremsstrahlung absorption and
Fresnel absorption were considered in these models. Surface tension, buoyancy, droplet
impact force and recoil pressure were considered to calculate the melt flow patterns. In the
following, fundamental physics, especially transport phenomena involved in hybrid laser-
arc welding will be elaborated.
2. Fundamentals of hybrid laser-arc welding
Since hybrid laser-arc welding involves laser welding, arc welding and their interactions as
well, complicated physical processes like metal melting and solidification; melt flow;
keyhole plasma formation; arc plasma formation and convection are typically involved,
which results in very complex transport phenomena in this welding process
40
. As known,

Hybrid Laser-Arc Welding 7
transport phenomena in welding, such as heat transfer; melt flow; and plasma flow, can
strongly affect both metallurgical structures and mechanical properties of the weld
41-45
. In
the following, transport phenomena in hybrid welding will be discussed and particular
attentions are given to 1) arc plasma formation and its effect on metal transfer and weld pool
dynamics; 2) laser-induced plasma formation and laser-plasma interaction; 3) recoil pressure
and other possible mechanisms contributing to keyhole formation and dynamics; 4) the
interplay among various process parameters; and 5) plasma - filler metal - weld pool
interactions.
Due to the different natures of heat and mass transfer mechanisms in metal and plasma,
different models are developed to study the fundamental physics in hybrid laser-arc
welding. One is for the metal region containing base metal, electrode, droplets, and arc
plasma. The other is for the keyhole region containing laser-induced plasma. There is a free
surface (liquid/vapor interface) separating these two regions. For the metal region,
continuum formation is used to calculate the energy and momentum transport
40
. For the
keyhole plasma region, laser-plasma interaction and the laser energy absorption mechanism
will be discussed. These two regions are coupled together and the VOF technique is used to
track the interface between these two regions
40
.
2.1. Transport phenomena in metal (electrode, droplets, and workpiece) and arc
plasma
Differential equations governing the conservation of mass, momentum and energy based on
continuum formulation are given below
46
:
Conservation of mass

( ) ( )
0
t
c
+V =
c
V (1)
where t is the time, is the density, and V is the velocity vector.
Conservation of momentum

2
0.5
( ) ( ) ( ) ( )
( )
l
l s s s
l l
l
s l r s
x
l
u p C
u u u u u u u u u
t x K
K
f f u u
c

c

| |
c
+V = V V
|
|
c
\ .
| | | |
V +V V + | |
| |
\ . \ .
r
V
V
J B
(2)

2
0.5
( ) ( ) ( ) ( )
( )
l
l s s s
l l
l
s l r s
y
l
u p C
v v v v v v v v v
t y K
K
f f v v
c

c

| |
c
+V = V V
|
|
c
\ .
| | | |
V +V V + | |
| |
\ . \ .
r
V
V
J B
(3)

Welding Processes 8

2
0.5
0
( ) ( ) ( ) ( )
( ) ( )
l
l s s s
l l
l
s l r s T drag
z
l
u p C
w w g w w w w w w w
t z K
K
f f w w g T T F
c

c

| |
c
+V = +V V
|
|
c
\ .
| | | |
V +V V + + + | |
| |
\ . \ .
J B
r
V
V
(4)
where u, v and w are the velocities in the x-, y- and z-directions, respectively, and Vr is the
relative velocity vector between the liquid phase and the solid phase. J is the current field
vector and B is the magnetic field vector. The subscripts s and l refer to the solid and
liquid phases, respectively; Subscript 0 represents the reference conditions; p is the
pressure; is the viscosity; f is the mass fraction; K, the permeability, is a measure of the
ease with which fluid passes through the porous mushy zone; C is the inertial coefficient;
|T is the thermal expansion coefficient; g is the gravitational acceleration; and T is the
temperature.
Conservation of energy

( )
( ) ( ) ( )
5
| |
( )( )
2
s
p p
b
l R
e p
k k
h h h h h
t c c
k
h
h h S
e c
c

c

o
| | | |
| | +V = V V V V
| |
\ . \ .
V
V + +
V
J
V - V J
s
(5)
where h is the enthalpy, k is the thermal conductivity, and cp is the specific heat. The first
two terms on the right-hand side of Eq. (5) represent the net Fourier diffusion flux. The third
term represents the energy flux associated with the relative phase motion. e is the electrical
conductivity; SR is the radiation heat loss; kb is the Stefan-Boltzmann constant; and e is the
electronic charge.
The third and fourth terms on the right-hand side of Eqs. (2)-(4) represent the first and
second order drag forces of the flow in the mushy zone. The fifth term represents an
interaction between the solid and the liquid phases due to the relative velocity. The second
term on the right hand side of Eq. (5) represents the net Fourier diffusion flux. The third
term represents the energy flux associated with the relative phase motion. All these
aforementioned terms in this paragraph are zero except in the mushy zone. In addition, the
solid phase is assumed to be stationary (VS= 0).
Conservation of species
( ) ( ) ( ) ( ) ( )
( ) ( ) ( )
l l
f f D f D f f f f
t
o o o o o o o
c

c
+V = V V V V V V V - V
s
(6)
where D is a mass diffusivity and f
o
is a mass fraction of constitute. Subscript, l and s,
represents liquid and solid phase respectively.

Hybrid Laser-Arc Welding 9
2.2. Transport phenomena in laser induced plasma
The vapor inside the keyhole is modeled as a compressible, inviscid ideal gas. No vapor
flow is assumed in the keyhole and the energy equation is given in the following form
47
:

r ,1
,1 ,
1
( ) ( q ) (1 )
(1 ) (1 ) (1 )
v
v v v pl laser iB
v
n
pl laser iB Fr iB mr
mr
k
h h k I
t c
k I
c
o
c
o o o
=
| |
= V V +V + +
|
|
\ .
+

(7)
where hv and v represent the enthalpy and density of the plasma; kv and cv represent the
thermal conductivity and specific heat of the plasma. The first term on the right-hand side of
Eq. (7) represents the heat conduction term. The second term represents the radiation heat
term and qr stands for the radiation heat flux vector. The fourth term represents energy
input from the original laser beam. The last term represents the energy input from multiple
reflections of the laser beam inside the keyhole.
2.3. Electrical potential and magnetic field
Arc plasma from GMA welding will not only provide heat to the base metal, but will also
exert magnetic force on the weld pool. The electromagnetic force can be calculated as
follows
48
:
Conservation of current

= 0 (8)
=

(9)
where | is the electrical potential. According to Ohm's law, the self-induced magnetic field
B
u
is calculated by the following Ampere's law:

0
z
r
B j rdr
o
r
u

}
=
(10)
where 0 = 4t x 10
-7
H m
-1
is the magnetic permeability of free space. Finally, three
components of the electromagnetic force in Eqs. (2)-(4) are calculated via

a
z
x
x x
B j
r
u

= J B
(11)

z
y
y
B j
r
u
= J B
(12)

r
z
B j
u
= J B
(13)

Welding Processes 10
2.4. Arc plasma and its interaction with metal zone (electrode, droplets, and weld
pool)
In welding, shielding gas is ionized and forms a plasma arc between the electrode and
workpiece. In the arc region, the plasma is assumed to be in local thermodynamic equilibrium
(LTE)
49
, implying the electron and the heavy particle temperatures are equal. On this basis, the
plasma properties, including enthalpy, specific heat, density, viscosity, thermal conductivity
and electrical conductivity, are determined from an equilibrium composition calculation
49
. It is
noted that the metal vaporized from the metal surface may influence plasma material
properties, but this effect is omitted in the present study. It is also assumed that the plasma is
optically thin, thus the radiation may be modeled in an approximate manner by defining a
radiation heat loss per unit volume
49
. The transport phenomena in the arc plasma and the
metal are calculated separately in the corresponding arc domain and metal domain, and the
two domains are coupled through interfacial boundary conditions in each time step.
Heat transfer
At the plasma-electrode interface, there exists an anode sheath region
49
. In this region, the
mixture of plasma and metal vapor departs from LTE, thus it no longer complies with the
model presented above. Since the sheath region is very thin, it is treated as a special
interface to take into account the thermal effects on the electrode. The energy balance
equation at the surface of the anode is modified to include an additional source term, Sa,
50,51

for the metal region.

(14)
The first term on the right-hand side of Eq. (14) is the contribution due to heat conduction from
the plasma to the anode. The symbol keff represents the thermal conductivity taken as the
harmonic mean of the thermal conductivities of the arc plasma and the anode material. is the
length of the anode sheath region. Tarc is the arc temperature and Ta is the temperature of the
anode. The second term represents the electron heating associated with the work function of
the anode material. Ja is the current density at the anode and

is the work function of the


anode material. The third term qrad is the black body radiation loss from the anode surface. The
final term qevap is the heat loss due to the evaporation of electrode materials.
Similar to the anode region, there exists a cathode sheath region between the plasma and the
cathode. However, the physics of the cathode sheath and the energy balance at the
nonthermionic cathode for GMA welding are not well understood
50-56
. The thermal effect
due to the cathode sheath has been omitted in many models and reasonable results were
obtained
50-54
. Thus, the energy balance equation at the cathode surface will only have the
conduction, radiation, and evaporation terms.

(15)
where keff is the effective thermal conductivity at the arc-cathode surface taken as the
harmonic mean of the thermal conductivities of the arc plasma and the cathode material. is
the length of the cathode sheath. Tc is the cathode surface temperature.

Hybrid Laser-Arc Welding 11
Force balance
The molten part of the metal is subjected to body forces, such as gravity and electromagnetic
force. It is also subjected to surface forces, such as surface tension due to surface curvature,
Marangoni shear stress due to temperature difference, and arc plasma shear stress and arc
pressure at the interface of arc plasma and metal. For cells containing a free surface, surface
tension pressure normal to the free surface can be expressed as
57

= (16)
where is the surface tension coefficient and is the free surface curvature.
The temperature-dependent Marangoni shear stress at the free surface in a direction
tangential to the local free surface is given by
58

(17)
where s is a vector tangential to the local free surface.
The arc plasma shear stress is calculated at the free surface from the velocities of arc plasma
cells immediately adjacent the metal cells

(18)
where is the viscosity of arc plasma.
The arc pressure at the metal surface is obtained from the computational result in the arc
region. The surface forces are included by adding source terms to the momentum equations
according to the CSF (continuum surface force) model
57
. Using F of the VOF function as the
characteristic function, surface tension pressure, Marangoni shear stress, arc plasma shear
stress, and arc pressure are all transformed to the localized body forces and added to the
momentum transport equations as source terms for the boundary cells. Based on these
assumptions, Hu et al. has successfully simulated the arc and droplet formation and effects
of current density and the type of shielding gas on arc formation in a GMA welding process,
as shown in Fig. 3.

Figure 3. Arc formation in a GMA welding process.

Welding Processes 12
2.5. Laser-induced recoil pressure and keyhole dynamics
In the laser welding process, the laser beam is directed to the metal surface, which first melts
the material and produces a small molten pool in the workpiece. The liquid metal is then
heated to high temperatures resulting in large evaporation rates. The rapid evaporation
creates a large recoil pressure on the surface of the molten layer depressing it downwards.
Thus, a cavity with large depth-to-width ration called keyhole is formed. Many investigators
believe that the balance between the recoil pressure and surface tension force determines the
shape of the keyhole. So, understanding the formation and behavior of the recoil pressure
becomes very important for studying the laser welding process. The recoil pressure results
from the rapid evaporation of the liquid metal surface. When the liquid metal on the surface
is heated to its boiling point, evaporation begins to occur. There is a very thin layer called
Knudsen layer adjacent to the liquid surface where the vapor escaping from the liquid
surface is in a state of thermodynamic non-equilibrium, i.e., the vapor molecules do not
have a Maxwellian velocity distribution. This occurs when the equilibrium vapor pressure
(i.e., the saturation pressure) corresponding to the surface temperature is large compared to
the ambient partial pressure of the vapor. Under these conditions the vapor adjacent to the
surface is dominated by recently evaporated material that has not yet experienced the
molecular collisions necessary to establish a Maxiwellian velocity distribution. The Knudsen
layer is estimated to be a few molecular mean free paths thick in order to allow for the
molecular collisions to occur that bring the molecules into a state of translational
equilibrium at the outer edge of the Knudsen layer. The flow field around the Knudsen
layer is shown in Fig. 4.

Figure 4. A schematic of the gas dynamic of vapor and air away from a liquid surface at elevated
temperature.
Anisimov
59
and Knight
60
did the early investigations on the Knudsen layer. Here a kinetic
theory approach
61
is used in the present study. The analysis proceeds by constructing an
approximate molecular velocity distribution adjacent to the liquid surface. Equations
describing the conservation of mass, momentum and energy across the Knudsen layer are
developed in terms of this velocity distribution. This gives Eqs. (19) and (20), as given
below, for gas temperature,
K
T , and density,
K
, outside of the Knudsen layer as functions
of the liquid surface temperature and the corresponding saturation density,
sat
.

Hybrid Laser-Arc Welding 13

2
2
1 1
1
1 2 1 2
K
L
T m m
T

t t

(
| |
(
= +
|
(
+ +
\ .
(

(19)

2 2
2
1 1
( ) 1 ( )
2 2
m m K L L
sat K K
T T m
m e erfc m me erfc m
T T

t
( | |
(
= + +
( |
(

\ .
(20)
The quantity, m, is closely related to the Mach number at the outer edge of the Knudsen
layer,
K
M , and is defined as, / 2 2 /
K V K K V
m u R T M = = , where
V
and
V
R are the
ratio of specific heats and the gas constant for the vapor, respectively. The value of m
depends on the gas dynamics of the vapor flow away from the surface. The gas
temperature, pressure and density throughout the vapor region (outside of the Knudsen
layer) are uniform. The contact discontinuity, that is, the boundary between vapor and air,
is an idealization that results due to the neglect of mass diffusion and heat conduction.
The velocity and pressure are equal in these regions,
K S
u u = and
K S
P P = , where the
subscript, S, denotes properties behind the shock wave. Note that, in general,
K S
T T = and
K S
= .
The thermodynamic state and velocity of the air on each side of the shock wave are related
by the Rankine-Hugoniot relations, where the most convenient forms to this application are
given by Eqs. (21) and (22).
K
M is the Mach number in the vapor, / 2
K K V V K
M u R T = .

2
1 1
1 1
4 4
S V V K V V K V V K
K K K
P R T R T R T
M M M
P R T R T R T




(
| |
+ +
(
| = + + +
(
|
\ . (

(21)

1 1
1 /
1 1
S S S S
T P P P
T P P P



| | | |
+ +
= + +
| |
| |

\ . \ .
(22)
The saturation pressure,
sat
P , is obtained from Eq. (23), where A, B and C are constants
which depend on the material. This is used to obtain the saturation density,
( )
/
sat sat V L
P R T = , assuming an ideal gas.

( ) ( )
log log
sat L
L
A
P B T C
T
= + (23)
Eqs. (20)-(23) are solved as a function of
L
T using an iterative solution method. The vapor
was assumed to be iron in the form of a monatomic gas with a molecular weight of 56, and
1.67
V
= . Quantities of particular interest are the recoil pressure,
r
P , and rate of energy loss
due to evaporation,
e
q , and they are given below.

Welding Processes 14

2
,
r K K K e V K K
P P u q H u = + =
(24)
2.6. Laser-plasma interaction and multiple reflections of laser beam in keyhole
In the keyhole, the laser beam is reflected and absorbed multiple times on the keyhole wall.
Each time when the laser beam travels inside the keyhole, it will interact with the keyhole
plasma. Multiple reflections of the laser beam and its absorption mechanism are critical in
determining the energy transfer in laser welding, which are discussed below.
Inverse Bremsstrahlung (IB) absorption
With the continuous heating of the laser beam, the temperature of the metal vapor inside the
keyhole can reach much higher than the metal evaporation temperature, resulting in strong
ionization, which produces keyhole plasma. The resulting plasma absorbs laser power by
the effect of Inverse Bremsstrahlung (IB) absorption. Eqs. (25) and (26) define the IB
absorption fraction of laser beam energy in plasma by considering multiple reflection
effects
62
:

0
,1
0
1 exp
s
iB pl
k ds o
| |
|
=
|
\ .
}
(25)

,
0
1 exp
m
s
iB mr pl
k ds o
| |
|
=
|
\ .
}
(26)
here,
,1 iB
o is the absorption fraction in plasma due to the original laser beam;
, iB mr
o is the
absorption fraction due to the reflected laser beam.
0
0
s
pl
k ds
}
and
0
m
s
pl
k ds
}
are, respectively, the
optical thickness of the laser transportation path for the first incident and multiple
reflections, and kpl is the plasma absorption coefficient due to inverse Bremsstrahlung
absorption
63
:

0.5
2 6
3 3 2
0
2
1 exp
2
6 3
e i e
pl
e e
e
n n Z e m
k g
kT kT
m ch m
t
e
t
c e
( | | | |
=
( | |
| |
(
\ . \ .
(27)
where Z is the average ionic charge in the plasma, e is the angular frequency of the laser
radiation,
0
c is the dielectric constant, k is the Boltzmanns constant, ne and ni are particle
densities of electrons and ions, h is Plancks constant, me is the electron mass, Te is the
excitation temperature, c is the speed of light, and g is the quantum mechanical Gaunt
factor. For the weakly ionized plasma in the keyhole, the Saha equation
63
can be used to
calculate the densities of plasma species:

Hybrid Laser-Arc Welding 15

( )
1.5
3
0 0
2
exp
e e
e i e i i
e
m kT
n n g g E
n g kT
h
t | |
=
|
|
\ .
(28)
Fresnel absorption
As discussed before, part of the laser energy will be absorbed by keyhole plasma and part of
the laser energy can reach the keyhole wall directly. So, the energy input (
laser
q ) for the
keyhole wall consists of two parts: 1) Fresnel absorption of the incident intensity directly
from the laser beam (
,Fr
I
o
) and 2) Fresnel absorption due to multiple reflections of the beam
inside the keyhole (
,mr
I
o
).

, , laser Fr mr
q I I
o o
= +
(29)

, ,1 1
(1 ) ( )
Fr laser iB Fr
I I
o
o o =
(30)

, ,1 ,
1
(1 ) (1 ) (1 ) ( )
n
mr laser iB Fr iB mr Fr mr
mr
I I
o
o o o o
=
=

(31)
where
laser
I is the incoming laser intensity. We assume the laser beam has in the simplest
case a Gaussiam-like distribution:

2
2
0
2
0
2
( , , ) exp
f
laser
f
f
r
r
I x y z I
r
r
| |
| |
|
| =
| |
\ .
\ .
(32)
where rf is the beam radius and rf0 is the beam radius at the focal position; I0 is the peak
intensity. Fr is the Fresnel absorption coefficient and can be defined it in the following
formula
64
:

2 2 2
2 2 2
1 1 (1 cos ) 2 cos 2cos
( ) 1
2
1 (1 cos ) 2 cos 2cos
Fr

c c c

c c c
| |
+ +
= + |
|
+ + + +
\ .
(33)
where is the angle of incident light with the normal of keyhole surface, n is the total number
incident light from multiple reflections, I

is the unit vector along the laser beam radiation


direction and n

is unit vector normal to the free surface. c is a material-dependent coefficient.


2.7. Radiative heat transfer in laser-induced plasma
When an intense laser beam interacts with metal vapor, a significant amount of the laser
radiation is absorbed by the ionized particles. The radiation absorption and emission by the
vapor plume may strongly couple with the plume hydrodynamics. This coupling, shown on
the right-hand side of Eq. (7), will affect the plasma laser light absorption and radiation
cooling terms. The radiation source term ( ) V
r
q is defined via

Welding Processes 16

4
(4 )
a b
k I Id
t
t V = O
} r
q
(34)
where ka, Ib and O denote the Planck mean absorption coefficient, blackbody emission
intensity and solid angle respectively. For the laser-induced plasma inside the keyhole, the
scattering effect is not significant compared with the absorbing and emitting effect. So it will
not lead to large errors to assume the plasma is an absorbing-emitting medium. The
radiation transport equation (RTE) has to be solved for the total directional radiative
intensity I
65
:

( ) ( ) ( ( ))
a b
I k I I V = s r, s r, s
(35)
where s and r denote a unit vector along the direction of the radiation intensity and the
local position vector. The Planck mean absorption coefficient is defined in the
following
65
:

1.5
0.5
2 6
3 3.5
128
27
e i
a
e
v
n n Z e g
k k
m
h c T
t
o
| |
| |
=
|
|
|
\ .
\ .
(36)
where ni and ne represent the particle density of ions and electrons, Tv is the temperature of
the plasma, Z stands for the charge of ions, e is the proton charge and me is the mass of
electrons.
2.8. Tracking of free surfaces
The algorithm of volume-of-fluid (VOF) is used to track the moving free surface
48
. The fluid
configuration is defined by a volume of fluid function, F(x,y,z,t), which is used to track the
location of the free surface. This function represents the volume of fluid per unit volume
and satisfies the following conservation equation:

( ) 0
dF F
F
dt t
c
= + V =
c
V
(37)
When averaged over the cells of a computing mesh, the average value of F in a cell is equal
to the fractional volume of the cell occupied by the fluid. A unit value of F means a cell full
of fluid and a zero value indicates a cell containing no fluid. Cells with F values between
zero and one are partially filled with fluid and identified as surface cells.
3. Results and discussions
Based on the aforementioned scientific principles governing the hybrid laser-arc welding
process, Zhou et al.
40,69,70
have successfully developed mathematical models to simulate the
transport phenomena like heat and mass transfer, melt flow; energy transport in keyhole
plasma, etc. in both pulsed and three-dimensional moving hybrid laser-MIG welding.
Detailed discussions are given in the following sections.

Hybrid Laser-Arc Welding 17
3.1. Two-dimensional hybrid laser-MIG welding
In this study, the base metal is assumed to be a stainless steel 304 containing 100 ppm of sulfur.
The laser energy is assumed to be in the Gaussian distribution and the divergence of the laser
beam is negligible because the focus length of the laser beam is less than 3mm. The laser
power and beam radius at the focus are 1800 W and 0.2 mm respectively. The laser power is
turned on at t = 0 s and shut down at t = 15 ms. To simulate the MIG process, droplet is
assumed to be spherical and is generated in a steady manner. The diameter of filler droplet is
assumed to be 0.35 mm, its initial speed is 0.5 m/s right above the weld surface, its initial
temperature is 2400 K, and its generation frequency is 1000 Hz. The droplet is assumed to be
fed into the keyhole from the top. Droplet generation and formation are actually related to
wire size and wire feed speed. Further information can be found in Ref. [13].
Fig. 5 shows the comparison of the cross-sectional view of a hybrid laser-MIG weld and a
laser weld. As shown, there is a "pore" in the laser weld, which is due to the rapid
solidification in laser welding. Detailed discussion on the formation of porosity in the weld
can be found in Ref. [14]. It is also noticed that there are some "undercuts" near the top edge
of the laser weld which is one of the major disadvantages of laser welding. In hybrid laser-
MIG welding, the reason why there is no pore found in the final weld was believed to be
mainly due to the addition of filler metal in the process. The momentum and energy carried
by the filler droplets greatly impact the fluid flow and heat transfer in the weld pool and the
shape of the solidified weld pool as well. The overall effect depends on the droplet size,
droplet generation frequency and droplet generation duration as well. With an optimal
operation window, a weld with desired shape and quality can be achieved in hybrid laser-
MIG welding. In addition, it is found that the additional heat input from the arc in hybrid
laser welding is transferred to the weld pool mainly in the region near the top of the weld,
which makes the top portion of the weld wider than that in laser welding. It is further found
the undercuts frequently observed in laser welding are eliminated and the shape of the final
weld can be modified by the extra filler metal coming from the MIG process. However, the
penetration depth in hybrid welding is noticed to be almost the same as that in laser
welding, which means the penetration depth in hybrid laser-arc welding mainly depends on
the laser power used, but not the arc power.

Figure 5. Comparison of weld bead shape between laser welding and hybrid laser-MIG welding.

Welding Processes 18
3.2. Interaction between filler droplets and weld pool
Fig. 6 shows typical interactions between droplets and weld pool in hybrid laser-arc
welding. The corresponding distributions of temperature, sulfur concentration, and melt
flow velocity are given in Figs. 7, 8 and 9, respectively. Since only the interaction between
filler droplets and weld pool is concerned in this discussion, the keyhole formation process
is ignored which can be found in Ref. [14]. As shown in Fig. 6, after the laser is shut off at t =
15.0 ms, the laser-induced recoil pressure decreases quickly. Under the action of surface
tension and hydrostatic pressure, the molten metal near the keyhole shoulder has tendency
to "fill up" the keyhole. At about t = 17.5 ms, the first droplet impinges onto the liquid metal
at the bottom of the keyhole. The downward momentum carried by the droplet causes the
droplet liquid to flow downward and outward along the keyhole wall, which can be seen
clearly by the sulfur composition shown in Fig. 8. Under the action of hydrostatic force and
surface tension, the liquid along the keyhole wall has a tendency to flow downward along
the keyhole wall. So the upward flow caused by the filler droplet impingement will be
weakened. So when the subsequent droplets falls into the keyhole, the liquid level in the
center of the keyhole rises, as shown in Fig. 6 at t = 21.5 ms. For the first several droplets, the
filler metal mainly diffuses along the longitude direction. Only the first droplet can spread
out along the solid-liquid interface driven by the downward momentum. However, as more
and more droplets impinge into the weld pool, a vortex is created, which helps the filler
metal to diffuse outwards in the latitude direction, as shown in Fig. 8.

Figure 6. Droplet and weld pool interaction in hybrid laser welding.

Hybrid Laser-Arc Welding 19

Figure 7. The corresponding temperature distributions as shown in Fig. 6.

Figure 8. The corresponding sulfur concentration distributions as shown in Fig. 6

Welding Processes 20
As shown in Fig. 9, there is an anticlockwise vortex in the middle waist of the keyhole. As
mentioned before, the liquid on the shoulder of the keyhole has a tendency to fill back along
the keyhole wall. When the droplets impinge into the keyhole, the outer liquid layer along
the keyhole wall has the same flow direction as the filler metal. So the flow direction of the
liquid metal here remains downward. Since the liquid is incompressible, the downward
flow will push up some amount of liquid upward. The kinetic energy of the fluid flow in the
center will be transferred into the potential and kinetic energy of the outward flow. So the
downward momentum becomes smaller and smaller and finally it changes its direction. As
shown in Fig. 9 at t = 21.5 ms, the flow direction changes from downward to outward at the
bottom of the vortex and then bounds upward on the solid keyhole wall. During the
upward flow process along the solid-liquid interface of the keyhole wall, the kinetic energy
is transferred into the potential energy and the velocity becomes smaller and smaller. Finally
the flow direction is changed to be inward by the back-filling momentum from the liquid on
the shoulder of the keyhole. As droplets continue to drip into the keyhole, more and more
downward momentum is added into the center of the keyhole, the vortex affected zone is
enlarged and the strength of the vortex is enhanced, which helps the filler metal to distribute
outward along with the vortex flow, as shown in Fig. 8. At t = 24.5 ms, the diffusion zone of
filler metal is much larger compared with that at t = 21.5 ms. Since the latitude diffusion of
filler metal has a close relationship with the vortex, the evolution of the vortex can be
deduced from the shape of the diffusion zone of the filler metal in the final fusion zone.
Moreover, at t = 24.5 ms, the downward velocity of the liquid in the center is quite large, the
mass from droplets is not enough to compensate the downward mass flow in the center of
the keyhole, which leaves the liquid surface decrease here.
After t = 25.0 ms, no droplet will be added into the keyhole. The fluid near the center of the
keyhole is bounced back under the action of hydrostatic force and surface tension force. As
shown at t = 29.0 ms in Fig. 9, the liquid in the keyhole starts to flow inward and downward,
which causes the size of the keyhole to become smaller and smaller. Finally the keyhole will be
filled, as shown at t = 49.0 ms in Fig. 6. During the backfill process, the vortex becomes weaker
and weaker. So the diffusion of filler metal is not improved much in the latitude direction,
which can be found by comparing those figures at t = 29.0 ms and at t = 46.0 ms in Fig. 5.9.
Moreover, from the distribution of filler metal at t = 46.0 ms as shown in Fig. 8, it can be
concluded that during the backfill process, majority of the filling metal comes from the upper
shoulder of the keyhole because only a little of the filler metal is located near the center of the
keyhole, which is brought here by the bouncing flow. As shown in Fig. 7, the filler droplet also
brings some heat into the weld pool, which will delay the solidification process. Since the
diffusion of filler in the fusion zone is greatly limited by the solidification, the delayed
solidification will give more time for the filler to diffuse. After the termination of droplets, the
heat input carried by droplets also decreases. Due to heat loss to the base metal through
conduction and to the surroundings through radiation and convection, the size of the molten
pool becomes smaller and smaller as a result of solidification. At t = 46.0 ms, the melt flow in
the weld pool is almost diminished and the temperature distribution is more uniform than
before, as shown in Figs. 9 and 7, respectively. The shape and composition of the weld will not
change much comparing with the completely solidified one.

Hybrid Laser-Arc Welding 21

Figure 9. The corresponding velocity distributions as shown in Fig. 6.
3.3. Modification of composition by adding filler metal
Since crack sensitivity of the weld is believed to be strongly related with the composition of
the weld pool, adding filler metal with anti-crack elements into the weld pool in hybrid laser
welding can thus improve the weld bead quality. However, the effect depends greatly on
the diffusion process in the weld pool. In the following, the effects of factors such as droplet
size, droplet generation frequency, impingement velocity of the droplet and its lasting
duration on the diffusion process are discussed by changing the condition of one specific
parameter, while keeping the rest of the parameters unchanged. If not specially mentioned,
the welding condition is defined as follows: the droplet diameter is 0.35 mm, its initial
velocity is 0.5 m/s, the generation frequency is 1000 HZ and the duration of droplet feeding
is 10.0 ms which starts at t = 15.0 ms and ends at t = 25.0 ms.

Welding Processes 22
3.3.1. Effect of droplet size on the diffusion process in hybrid laser welding
Three studies are carried out with a droplet size of 0.3 mm, 0.35 mm and 0.4 mm
respectively. As shown in Fig. 10, with the increase of droplet size, the latitude diffusion of
filler metal is enlarged. From the previous discussions on the diffusion process, the latitude
diffusion of the filler metal is found to be closely related to the vortex in the weld pool. The
strength and the affected zone of the vortex depend on the downward momentum carried
by droplets, which is the product of droplet mass and velocity. As the droplet size increases,
the downward momentum increases, which will lead to a stronger vortex. So the diffusion
zone is enlarged outward, especially in the middle depth of the keyhole where the vortex is
located. This is clear shown by comparing those figures for d = 0.30 mm and d = 0.35 mm in
Fig. 10. Meanwhile, larger downward momentum from larger droplet also leads to a strong
bouncing flow near the center of keyhole after termination of droplet feeding, which helps
filler metal to diffuse into the upper layer in the final weld, as shown in Fig. 10 for d = 0.40
mm. Moreover, larger droplet size brings more filler metal into the keyhole. The heat input
carried by droplets also increases, which helps delay the solidification of the fusion zone.
Thus, the filler metal has more time to diffuse into the weld pool before its solidification.
More filler metal also helps to increase the filler concentration in the final weld. However,
larger droplets also lead to some negative effects on the diffusion of filler metal near the
center of the weld zone. After the termination of droplet feeding, the melt surface near the
center of the keyhole will continue to go down due to the larger hydrodynamic pressure
caused by the downward momentum. This will lead to a deep hole there. During the
backfill process of this hole, some metal from the upper part of the keyhole may flow into
the bottom part of this hole. Since the concentration of filler metal in the upper part of
keyhole is very low, it leaves a low diffusion zone of filler metal in the center of the final
weld, as shown in Fig. 10 for d = 0.40 mm.

Figure 10. Effect of droplet size on diffusion process in hybrid laser welding.
3.4. Effect of droplet generation frequency on diffusion of filler metal in fusion
zone
In the hybrid laser welding process, the droplet is generated at a specific frequency that is
controlled by the wire feed rate. The effect of droplet generation frequency on diffusion of

Hybrid Laser-Arc Welding 23
filler metal in the fusion zone is shown in Fig. 11. In the study, the droplet generation
frequency is 500 HZ, 667 HZ and 1000 HZ, which corresponds to the generation of one
droplet every 2.0 ms, 1.5 ms and 1.0 ms. As shown, the diffusion of filler metal in the weld
pool is improved with the increase of generation frequency. This can be interpreted through
the above analysis on the interaction between the droplets and weld pool. As mentioned
before, the latitude diffusion of filler is mainly through the vortex flow induced by the
impingement of droplets. With the increase of generation frequency, more droplets fall into
the weld pool per unit time, which results in higher downward momentum per unit time. So
the vortex in the weld pool is enhanced, which helps the filler metal diffuse in the latitude
direction. Meanwhile, the total amount of filler metal added into the weld pool also
increases with higher generation frequency, which also helps increase the concentration of
filler metal in the final weld and increase the diffusion time, as mentioned before.
Furthermore, the longitude distribution of filler metal is found to be improved with higher
generation frequency. As shown, in the case f = 500 HZ, there exists a low filler metal
concentration zone in the lower part of the keyhole due to the weak strength of the vortex in
the weld pool and a long delay time between the droplet generation. When frequency
increases to 1000 HZ, the size of this zone is greatly reduced.

Figure 11. Effect of droplet generation frequency on diffusion process in hybrid laser welding.
3.5. Effect of droplet generation duration on diffusion of filler metal in fusion
zone
In hybrid laser welding, the termination of droplet generation can be achieved through
the control of removal of the filler wire. The effect of controlling the droplet generation
duration on metal diffusion in the weld pool is investigated. As shown in Fig. 12, three
cases are carried out with droplet generation duration at 5.0 ms, 10.0 ms and 15.0 ms,
respectively. For short duration of 5.0 ms, the vortex induced by the downward
momentum of the droplet is not completely developed because of lower downward
momentum, which leads to poor latitude distribution of filler metal. In this case, most of
the filler metal is located in the lower part of the keyhole. During the backfill process, the
bounced flow is not strong enough to push the filler metal upward to the upper part of
the keyhole. The keyhole is filled with the base metal liquid where no filler metal exists.
So the longitude filler diffusion is also poor with a short duration of droplet generation.

Welding Processes 24
With the increase of the duration length to 10.0 ms, more filler metal will fall into the
keyhole. The vortex in the weld pool is enhanced with the increased downward
momentum which improves the latitude diffusion. Meanwhile, the droplets are
distributed along the entire depth of the keyhole, which leads to better longitude
distribution of filler metal. Moreover, the total amount of filler metal also increases with
the increase of duration, which also helps the diffusion of filler metal in the fusion zone,
as mentioned before. So both the longitude and latitude diffusion of filler metal are
improved, as shown. However, with a further longer droplet generation to 15.0 ms, the
downward momentum is accumulated due to the continuous impingement of the droplets
into the weld pool, which leads to a deep hole in the weld pool. During the backfill
process of this hole, the filler metal is mainly located in the bottom, which cannot bounce
back in time before the base metal fluid from the upper shoulder arrives at the bottom of
this hole, which leaves a low diffusion zone of filler metal in the center of final weld, as
shown in Fig. 12.

Figure 12. Effect of droplet impingement duration on diffusion process in hybrid laser welding.
3.6. Three-dimensional hybrid laser-MIG welding
Fig. 13 shows a schematic sketch of a three-dimensional hybrid laser-MIG welding. In this
study, the laser power is 2.0 kW and the laser beam radius is 0.2 mm and the focal plane is on
the top surface of the base metal. The laser beam is started at x = 3.75 mm. The laser beam
begins to move after being held for 20.0 ms for the keyhole to reach a certain depth. The
welding speed is 2.5 cm/s and the arc power is 1 kW. Droplet begins to fall onto the base metal
at t = 20.0 ms and the radius of the droplet is 0.25 mm. The droplet feeding frequency is 86 Hz
and its initial speed is 30 cm/s. The distance between arc center and laser beam center is 1 mm.
Fig. 14 is the side-view (at Y = 0) of the hybrid laser welding process showing a sequence of a
droplet impinging onto the weld pool at different times. Fig. 15 shows the corresponding
sulfur concentration distribution during the hybrid welding process, indicating the mixing
process in the welding. Fig. 16 shows the corresponding velocity distributions in the weld
pool. As shown in Fig. 15, the filler droplet did not mix well with the base metal in this case.
Most of the droplet is just stacking on the top of the weld coupon and only small amount of
the filler metal is diffused into the base metal near the solid-liquid interface. The poor mixing
may have been caused by the relative long distance between the laser beam and MIG arc

Hybrid Laser-Arc Welding 25
center. The filler droplet is impinging into the weld pool where only a small amount of liquid
metal exists. Since the temperature of this part of liquid metal is low, due to the quick
solidification process there, the liquid metal there solidifies very quickly. The droplet flowing
downward does not have enough time to flow around and exchange the momentum and mix
with the base metal before it solidifies, as shown in Fig. 16. Therefore, most of the filler metals
are just stacking on top surface of the base metal.

Figure 13. Schematic sketch of 3-D hybrid laser keyhole welding.

Figure 14. A typical sequence showing the impinging process and temperature distributions in 3-D
moving hybrid laser keyhole welding.

Welding Processes 26

Figure 15. The corresponding sulfur concentration distributions as shown in Fig. 14.

Figure 16. The corresponding velocity distributions as shown in Fig. 14.

Hybrid Laser-Arc Welding 27
There are a lot of process parameters which can affect the mixing of filler droplets into the
weld pool in three-dimensional hybrid laser-arc welding. These include laser-arc distance,
laser and arc powers, welding speed, wire feed speed and filler droplet size, etc. In the
following study, the effect of laser-arc distance on diffusion is conducted by decreasing the
laser-arc distance to 0.6 mm. Fig. 17 shows the mixing process during the welding. As
shown, the droplet is now mixing with the base metal much better than in the previous case.
In this case, since the laser-arc distance is decreased, the filler droplet can impinge into a
region in the weld pool where there exists a lot of liquid metal with strong velocity and high
temperature. This strong velocity liquid metal flow will interact with the impinging
droplets, creating a strong momentum exchange between the droplets and weld pool, which
can force the droplet to flow in all directions. Hence, a better mixing can be achieved. Also,
in this case, there are more hot liquid metals in the droplet-weld pool interaction zone, thus
creating relatively longer time for the droplet to mix and diffuse into the base metal. Hence,
a better mixing of droplets into the weld pool is achieved.

Figure 17. Diffusion process in 3-D hybrid laser keyhole welding with shorter laser-arc distance.
4. Future trends
Although hybrid laser-arc welding has been under investigation and development and
gaining increasing acceptance in recent years, good understanding of the underlying
physics remains a challenge. For example, the interaction between the laser and the arc has

Welding Processes 28
been observed to enhance arc stability and push the arc towards the laser keyhole, resulting
in a deeper penetration. However, the origin of this synergistic interaction between the arc
and laser plasma is not well understood. Measuring the distributions of electron
temperatures and densities in the plasma can provide a better understanding of the laser-arc
interaction
9
. Porosity formation is believed to be strongly related to the keyhole collapse
process. Hence, better understanding of keyhole stability and dynamics through
experimental and theoretical studies would be beneficial. Hybrid welding is known to
produce welds with desirable widths and depths, but the maximum gap tolerance and weld
penetration for various welding conditions have not been quantified. In the future,
advanced mathematical modeling of the heat transfer and fluid flow will enable accurate
predictions of weld profile and cooling rates in the welding process, which is critical in
understanding the evolution of weld microstructures and residual stress formation in welds.
Thus, the hybrid welding process can be optimized to obtain quality welds with no
cracking, no brittle phase and less thermal distortion. Better sensing and process control of
the hybrid welding process would also be helpful in expanding its applications
67
.
Author details
J. Zhou
Department of Mechanical Engineering, Pennsylvania State University, The Behrend College, USA
H.L. Tsai
Department of Mechanical and Aerospace Engineering,
Missouri University of Science and Technology, USA
5. References
[1] Tusek, J., and Suban, M., Hybrid welding with arc and laser beam, Science and
Technology of Welding and Joining, 4(5), 308-311, 1999.
[2] Mahrle, A., and Beyer, E., Hybrid laser beam welding Classification, characteristics,
and applications, Journal of Laser Applications, 18(3), 169-180, 2006.
[3] Steen, W.M., Arc-augmented laser processing of materials J. Appl. Phys., 51, 56365641,
1980.
[4] Campana, G., Fortunato, A., Ascari, A., Tani, G., and Tomesani, L., The influence of arc
transfer mode in hybrid laser-MIG welding, J. Mater. Process. Technol., 191, 111113, 2007.
[5] Casalino, G., Statistical analysis of MIG-laser CO2 hybrid welding of AlMg alloy, J.
Mater. Process. Technol., 191, 106110, 2007.
[6] Hu, B., and Richardson, I.M., Microstructure and mechanical properties of AA7075(T6)
hybrid laser/GMA welds, Material Science Engineering A, 459, 94-100, 2007.
[7] Feng, Z.L., Processes and mechanisms of welding residual stress and distortion, Woodhead
Publishing Ltd, Cambridge, England, 2005.
[8] Metzbower, E.A., Denney, P.E., Moon, D.W., Feng, C.R., and Lambrakos, S.G., Thermal
analysis and microharness maping in hybrid laser welds in a structural steel, Materials
Science Forum, 426-432, 4147-4152, 2003.

Hybrid Laser-Arc Welding 29
[9] Ribic, B., Palmer, T.A., and Debroy, T., Problems and issues in laser-arc hybrid
welding, International Materials Reviews, 54(4), 223-244, 2009.
[10] G. Song, L. M. Liu and P. C. Wang: Material Science Engineering A, 429, 312319, 2006.
[11] Nonn, A., Dahl, W., and Bleck, W., Numerical modelling of damage behaviour of laser-
hybrid welds, Engineering Fracture Mechanics, 75, 32513263, 2008.
[12] Pinto, H., Pyzalla, A., Hackl, H., and Bruckner, J., A Comparative Study of
Microstructure and Residual Stresses of CMT-, MIG- and Laser-Hybrid Welds,
Materials Science Forum, 524525, 627632, 2006.
[13] Zhang, J. X., Xue, Y., and Gong, S. L., Residual welding stresses in laser beam and
tungsten inert gas weldments of titanium alloy, Science and Technology of Welding &
Joining, 10, 643646, 2005.
[14] Liu, L.M., Song, G., and Wang, J.F.,, Materials Science Forum, 488489, 361364, 2005.
[15] Liu, L.M., Song, G., Liang, G.L., and Wang, J.F, Pore formation during hybrid laser
tungsten inert gas arc welding of magnesium alloy AZ31B mechanism and remedy,
Material Science and Engineering A, 390, 76-80, 2005.
[16] Katayama, S., Uchiumi, S., Mizutani, M., Wang, J., and Fujii, K., Penetration and
porosity prevention mechanism in YAG laser-MIG hybrid welding, Welding
International, 21, 2531, 2007.
[17] Fellman, A., and Kujanpaa, V., The effect of shielding gas composition on welding
performance and weld properties in hybrid CO2 laser-gas metal arc welding of carbon
manganese steel, Journal of Laser Applications, 18, 1220, 2006.
[18] Kim, T., Suga, Y., and Koike, T., Welding of thin steel plates by hybrid welding process
combined TIG arc with YAG laser, JSME International Journal Series A, 46A, 202207,
2003.
[19] Kim, Y. P., Alam, N., and Bang, H. S., Observation of hybrid (cw Nd:YAG laser + MIG)
welding phenomenon in AA5083 butt joints with different gap conditions, Science and
Technology of Welding & Joining, 11(3), 295307, 2006.
[20] Murakami, K., Mitooka, Y., Hino, M., Iogawa, H., Ono, H., and Katayama, S., , J. Japn.
Inst. Met., 70, 134137, 2006.
[21] Qin, G. L., Lei, Z., and Lin, S. Y., Effects of Nd:YAG laser + pulsed MAG arc hybrid
welding parameters on its weld shape, Science and Technology of Welding & Joining,
12(2), 7986, 2007.
[22] Swanson, P. T., Page, C. J., Read, E., and Wu, H. Z., Plasma augmented laser welding of
6 mm steel plate , Science and Technology of Welding & Joining, 12(2), 153160, 2007.
[23] Eboo, M., Steen, W.M., and Clarke, J., Arc-augmented laser welding, Advances in
Welding Processes, Proc. of the 4th Int. Conf., Harrogate, UK, 257265, 1978.
[24] Steen, W. M., and Eboo, M., Arc augmented laser welding, Metal Construction, 11, 332
335, 1979.
[25] Bagger, C., and Olson, F.O., Review of laser hybrid welding, Journal of Laser
Applications, 17(1), 2-14, 2005.
[26] Cui, H., Decker, I., and Ruge, J., Wechselwirkungen zwischen WIGSchweilichtbogen
und fokussiertem Laserstrahl, Proc. of the Conference Laser89 , Springer, Berlin, 577581,
1989.

Welding Processes 30
[27] Finke, B. R., Stern, F., and Decker, I., Auswirkungen eines untersttzenden Laserstrahls
auf den WIG-Schweiprozess, DVS-Ber., 135, 149152, 1991.
[28] Cui, H., Decker, I., Pursch, H., Ruge, J., Wendelstorf, J., and Wohlfahrt, H.,
Laserinduziertes Fokussieren des WIG-Lichtbogens, DVS-Ber., 146, 139143, 1992.
[29] Dilthey, U., Lder, F., and Wieschemann, A., Expanded capabilities in the welding of
aluminium alloys with the laser-MIG hybrid process, Aluminium, 75, 6475, 1999.
[30] Graf, T., and Staufer, H., Laser hybrid process at Volkswagen, International Institute of
Welding, IIW-Doc. XII-1730-02, 1999.
[31] Jokinen, T., Viherv, T., Riikonen, H., and Kujanp, V., Welding of ship structural steel
A36 using a Nd:YAG laser and gas metal arc welding, Journal of Laser Applications, 12,
185188, 2000.
[32] Hfemann, M., Fersini, M., Szinyur, J., Haferkamp, H., and Cordini, P., Laser-GMA-
hybrid welding of zinc-coated steel for hydroforming applications, Proc. of the 1st Int.
WLT-Conf. on Lasers in Manufacturing, Munich, Germany, 466472, 2001.
[33] Dilthey, U., and Keller, H., Prospects in laser GMA hybrid welding of steel, Proc. of the
1st Int. WLT-Conf. on Lasers in Manufacturing, Munich, Germany, 453465, 2001.
[34] Waltz, C., Seefeld, T., and Sepold, G., Laser-GMA welding and its influence on bead
geometry and process stability, Proc. of the 1st Int. WLT-Conf. on Lasers in Manufacturing,
Munich, Germany, 444452, 2001.
[35] Shibata, K., Sakamoto, H., and Iwase, T., Laser-MIG hybrid welding of aluminium
alloys, Proc. of the 1st Int. WLT-Conf. on Lasers in Manufacturing, Munich, Germany, 436
443, 2001.
[36] Jokinen, T., Jernstrm, P., Karhu, M., Vanttaja, I., and Kujanp, V., Optimisation of
parameters in hybrid welding of aluminium alloy, Proc. Of the 1st Int. Symp. on High
Power Laser Macroprocessing, Proc. SPIE 4831, 307312, 2003.
[37] Ishide, T., Tsubota, S., and Watanabe, M., Latest MIG, TIG Arc-YAG laser hybrid
welding systems for various welding products, Proc. of the 1
st
Int. Symp. on High Power
Laser Macroprocessing, 2003 Proc. SPIE 4831, 347352,2003.
[38] Hyatt, C. V., Magee, K. H., Porter, J. F., Merchant, V. E., and Matthews, J. R., Laser-
assisted gas metal arc welding of 25-mm-thick HY-80 plate, Welding Journal
Symposium, Miami, FL, U. S.A, 163s172s, 2001.
[39] Fellman, A., Jernstrm, P., and Kujanp, V., The effect of shielding gas composition in
hybrid welding of carbon steel, Proceedings 9th NOLAMP, Conference on Laser Materials
Processing in the Nordic Countries, Trondheim, Norway, 103112, 2003.
[40] Zhou, J., and Tsai, H.L., Modeling of transport phenomena in hybrid laser-MIG
keyhole welding, International Journal of Heat and Mass Transfer, 51, 4353-4366, 2008.
[41] Page, C. J., Devermann, T., Biffin, J., and Blundell, N., Plasma augmented laser welding
and its applications , Science and Technology of Welding & Joining, 7(1), 110, 2002.
[42] S. M. Joo, Y. P. Kim, C. S. Ro, H. S. Bang and J. U. Park, , Adv. Nondestr. Eval., 270273,
23832388, 2004.
[43] Graf, T., and Staufer, H., Laser-hybrid welding drives VW improvements, Welding
Journal, 82, 4248, 2003.

Hybrid Laser-Arc Welding 31
[44] Ono, M., Shinbo, Y., Yoshitake, A., and Ohmura, M., Welding properties of thin steel
sheets by laser-arc hybrid welding: laser-focused arc welding, Proc. 1st Int. Symp. on
High-power laser macroprocessing, Osaka, Japan, 369374, 2002.
[45] Song, G., Liu, L. M., Chi, M. S., and Wang, J. F., Investigations of Laser-TIG Hybrid
Welding of Magnesium Alloys, Materials Science Forum, 488489, 371375, 2005.
[46] Chiang, K.C., Tsai, H.L., Shrinkage-induced fluid flow and domain change in two-
dimensional alloy solidification, Int. J. Heat and Mass Transfer, 35, 1763-1769, 1992.
[47] Zhou, J., Tsai, H.L., and Wang, P.C., Transport Phenomena and Keyhole Dynamics
during Pulsed Laser Welding, ASME Journal of Heat Transfer, 128(7), 680-690, 2006.
[48] Wang, Y., Modeling of three-dimensional gas metal arc welding process, Ph.D. Thesis,
University of Missouri-Rolla, 1999.
[49] Lowke, J.J., Kovitya, P., and Schmidt, H.P., Theory of free-burning arc columns
including the influence of the cathode, J. Phys. D: Appl. Phys., 25, 1600-1606, 1992.
[50] Haidar, J., A theoretical model for gas metal arc welding and gas tungsten arc
welding, J. Appl. Phys., 84, 3518-3529, 1998.
[51] Haidar, J., Lowke, J.J., Predictions of metal droplet formation in arc welding, J. Appl.
Phys. D: Appl. Phys., 29, 2951-2960, 1996.
[52] Haidar, J., An analysis of the formation of metal droplets in arc welding, J. Phys. D:
Appl. Phys., 1233-1244, 1998.
[53] Haidar, J., Prediction of metal droplet formation in gas metal arc welding. II., J. Appl.
Phys., 84, 3530-3540, 1998.
[54] Haidar, J., An analysis of heat transfer and fume production in gas metal arc welding.
III., J. Appl. Phys., 85, 3448-3459, 1998.
[55] Lowke, J.J., Kovitya, P., and Schmidt, H.P., Theory of free-burning arc columns
including the influence of the cathode, J. Phys. D: Appl. Phys., 25, 1600-1606, 1995.
[56] Zhu, P., Lowke, J.J., Morrow, R., A unified theory of free burning arcs, cathode sheaths
and cathodes, J. Phys. D: Appl. Phys., 25, 1221-1230, 1992.
[57] Brackbill, J.U., Kothe, D.B., and Zemach, C., A continuum method for modeling surface
tension, J. of Computational Physics, 100, 335-354, 1992.
[58] Fan, H.G., and Kovacevic, R., Droplet formation, detachment, and impingement on the
molten pool in gas metal arc welding, Metall. Trans., 30B, 791801, 1999.
[59] Anisimov, S.I., Vaporization of metal absorbing laser radiation, Soviet Physics, JETP 27
(1), 182-183, 1968.
[60] Knight, C.J., Theoretical modeling of rapid surface vaporization with back pressure,
AIAA J., 17(5) 519-523, 1979.
[61] Kanouff, M.P., Kassinos, A., Noble, D.R., and Schunk, P.R., Laser spot weld modeling
using an ALE finite element method, Sandia National Laboratories Report.
[62] Kaplan, A., A model of deep penetration laser welding based in calculation of the
keyhole profile, J. Phys. D: Appl. Phys., 27, 1805-1814, 1994.
[63] Duley, W., Laser welding, John Wiley & Sons, 1999.
[64] Ducharme, R., Williams, K., Kapadia, P., Dowden, J., Steen, B., and Glowacki, M., The
laser welding of thin metal sheets: an integrated keyhole and weld pool model with
supporting experiments, J. Phys. D: Appl. Phys., 27, 1619-1627, 1994.

Welding Processes 32
[65] Ho, J.R., Grigoropoulos, C.P., Humphrey, J.A.C., Gas dynamics and radiation heat
transfer in the vapor plume produced by pulsed laser irradiation of aluminum, J. Appl.
Phys., 79, 7205-7215, 1996.
[66] Bagger, C., Olsen, F.O., Review of hybrid laser welding, J. Laser Appl., 17, 2-14, 2005.
[67] Ribic, B., Palmer, T.A., DebRoy, T., Problems and issues in laser-arc hybrid welding,
Int. Mater. Rev., 54 (4), 223-244, 2009.
[68] Kong, F. and Kovacevic, R., 3D Finite Element Modeling of the Thermally Induced
Residual Stress in the Hybrid Laser/Arc Welding of Lap Joint, Journal of Materials
Processing Technology, 941-950, 2010.
[69] Zhou, J., Zhang, W. H., Tsai, H. L., Marin, S. P., Wang, P. C., and Menassa, R.,
Modeling the transport phenomena during hybrid laser-MIG welding process,
Proceedings IMECE03, ASME International Mechanical Engineering Congress & Exposition,
Washington, DC, IMECE2003-41763, 1-8, 2003.
[70] Zhou, J., and Tsai, H. L., Investigation of mixing and diffusion processes in hybrid spot
laser MIG keyhole welding, Journal of Physics D: Applied Physics, 42, 1-15, 2009.
Chapter 2




2012 Ozaki and Kutsuna, licensee InTech. This is an open access chapter distributed under the terms of
the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Dissimilar Metal Joining of Zinc Coated Steel and
Aluminum Alloy by Laser Roll Welding
Hitoshi Ozaki and Muneharu Kutsuna
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/48242
1. Introduction
Nowadays, the car industry has targets to improve fuel consumption due to the problems of
the global environment. For example, in the Corporate Average Fuel Economy, CAFE,
which was approved in the U. S. Congress in July 2011, attaining the average fuel efficiency
of 54.5 mpg (23.2 km/l) by 2025 is called for (Yamamoto, 2012). In addition, its obliged to
improve the average fuel efficiency to 35.5 mpg (15.1 km/l) by 2017. This is a significant
increase compared with the present 27.5 mpg (11.7 km/l) which have been fixed for the past
11 years. Hence, the further weight reduction of cars has become imperative.
On the other hand, the weight of vehicle which influences the fuel consumption directly is
increasing owing to a rise of safety awareness of automobile users, tightening of safety
standards, and diversification and sophistication of needs. Therefore, the car industry has
conflicting targets of the low fuel consumption by lightening car body and safety
improvement. Additionally these targets should be achieved in together. Then, a hybrid
body structure concept has been proposed.
In this concept, high strength steel and light alloy are arranged in the right places of the car
body. For instance, by using high strength steel for pillars, a lightening with improving
strength is achieved. In addition, by using aluminum alloys for the bonnet, the door panel,
and the trunk lid, a further lightening is achieved. In order to make this hybrid body
structure, the welding technology for joining of steel to aluminum with high reliability and
productivity is required. However, its difficult to join steel to aluminum by conventional
fusion welding.
Generally, in respect to dissimilar metal joining by the fusion welding, because of melting
and alloying largely both metals, intermetallic compound layer formed at joint interface
grows thick; oxide firm is formed; hot cracking is generated. Consequently, high joint

Welding Processes 34
strength can not be gotten. Moreover, when melting point of both metals has large
difference, burn-through occurs in the metal of lower melting point (Nishimoto et al., 2005).
In order to solve these problems, many studies of the dissimilar metal joining have been
conducted such as resistance welding (Okita, 2004), laser welding (Katayama, 2004),
explosive welding (Satou, 2004), friction stir welding (Okamura & Aota, 2004; Katoh &
Tokisue, 2004), and diffusion bonding (Ohashi, 2004). As a result, solid-phase bonding is
mainly put to practical use. However, higher productivity, joint strength and flexibility are
needed to expand the coverage of application of the dissimilar metal joints. Since laser
welding has advantages such as local heat input, short process time and high flexibility
compared with other welding processes, there has been considerable research on the
welding of steel and aluminum. In the recent, the studies of keyhole welding (Torkamany, et
al., 2010), laser welding-brazing (Dharmendra, et al., 2011), dual-beam YAG laser welding
(Yan, et al., 2010), and thermal cycle during laser welding (Fan, et al., 2011) are carried out.
Then, the welding process of steel and aluminum which have been regarded as difficult,
Laser Roll Welding has been developed for joining of dissimilar metals by M. Kutsuna, M.
Rathod and A. Tsuboi in 2002. This welding is a hybrid welding process, as shown in Fig.1,
combined high-temperature heating by a 2.4 kW CO2 laser with pressurizing by a pressure
roller. Its registered as Japanese Patents No. 3535152 and No. 3692135.

Figure 1. Schematic diagram of Laser Roll Welding process
Fig.2 shows Fe-Al phase diagram (Massalski, et al., 1986). In this figure, there are varioius
intermetallic compounds, hereafter, called as IMCs, and they are grouped as Fe-rich
compounds, FeAl and Fe3Al, and Al-rich compounds, FeAl2, Fe2Al5 and FeAl3. M. Yasuyama
et al. (1996) have shown the mechanical properties of these IMCs. Vickers hardness of
cast IMCs is shown in Table 1, and mechanical properties by compressive test are shown
in Fig.3.

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 35

Figure 2. Fe-Al binary equilibrium diagram
Table 1. Vickers hardness of Fe-Al intermetallic compounds

Figure 3. Stress-strain curves in compressive test of Fe-Al intermetallic compounds
While Al-rich IMCs are hard and brittle, Fe-rich IMCs show slight ductility and high
strength. These brittle IMCs constitute barriers to the dissimilar metal welding of steel to

Welding Processes 36
aluminum. Kutsuna et al. thought if the thickness of brittle IMC layer is minimized and the
formation of more ductile IMC's is promoted, high reliable joints are able to be obtained.
Therefore, they developed the Laser Roll Welding process. In this process, since the thermal
cycle for joining can be shortened by laser heating, the formation of the brittle IMCs can be
easily controlled. Furthermore, good contact of a steel sheet and an aluminum sheet and
rapid heat transfer from the steel sheet to the aluminum sheet are conducted by the pressure
roller.
Kutsuna et al. produced Laser Roll Welidng equipment experimentally, and conducted basic
studies of the welding of low carbon steel to aluminum alloy (Rathod &Kutsuna, 2003,
2004). As a result, when the IMC layer thickness was from 4 to 8 m, failure of base metal
sheet was obtained after tensile shear test of specimen.
Until now, its found that Laser Roll Welding has two types of welding mechanisms. The
combination of iron and aluminum is that of metals with large difference in their melting
points (Ozaki & Kutsuna, 2007). The combination of titanium and aluminum also
corresponds to this category (Ozaki et al., 2008). In this case, only the metal with a higher
melting point is heated up to an elevated temperature for avoid the formation of brittle
IMCs. For example, in the case of steel to aluminum joint, steel is heated up to 1200
o
C,
because the max formation temperature of Fe2Al5 is 1169
o
C. The interlayer is formed and
cooled rapidly to minimize the thickness of brittle IMCs.
The other is a combination of metals with eutectic reaction at a lower temperature than
melting point. For instance, the combination of titanium and iron falls into this type (Ozaki
et al., 2007). In this case, the eutectic reaction occurs at a lower temperature, about 1085
o
C,
than melting points of both metals. The interlayer of the eutectic phase, -Ti and TiFe, is
formed at the interface.
In this way, uncoated materials have been mainly used in the previous study of Laser Roll
Welding. However, coated materials such as zinc coated steel have not investigated enough
yet. In this chapter, Laser Roll Welding of zinc coated steel and 6000 series aluminum alloy
was conducted, and the weldability was investigated. Two types of zinc coated steel were
used. One is hot-dip galvanized steel called as GI, the other is hot-dip galvannealed steel
called as GA. The former is used mainly in Europe, the latter is in Japan. Then, the
influences of process parameters, such as welding speed and roll pressure, on the formation
of intermetallic compound layer and the change of zinc coated layer have been investigated
to get sound joints with these galvanized steel. Furthermore, the effects of process
parameters on joint performance have been also discussed.
2. Experimental procedure
2.1. Materials used
As materials, two types of zinc coated steel and A6000 series aluminum alloy, Al-0.5Mg-
1.0Si, were used for joining. The zinc coated steels were hot-dip galvanized steel, hereafter,

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 37
called as GI, and hot-dip galvannealed steel, hereafter, called as GA, respectively. The
dimension of the zinc coated steel sheets were 125 x 180 x 0.55 mm and that of the A6000
series aluminum alloy sheet, hereafter, called as A6000, was 125 x 180 x 1 mm. The zinc
coating weight of the GI sheet was 60 g/m
2
and the GA was 45 g/m
2
.
The surface of the steel sheets was coated with graphite by using a graphite spray to
increase the absorption rate of laser beam. The thickness of the coating layer was
approximately 10 m. The faying surface of the steel sheets was only degreased with
ethanol alcohol. The faying surface of the A6000 sheet was wire-brushed, polished by emery
paper #600 and cleaned by ethanol before welding. Then, to remove the oxide firm
on aluminum alloy sheet, the surface of the aluminum alloy was spread with flux,
KAlF4:K2AlF5H2O, 17-25 wt%, with particle size of 15 to 21 m.
The setup of specimens is shown in Fig.4. A zinc coated steel sheet is placed as a top plate;
an A6000 sheet is placed as a bottom with 3 mm width overlapping.

Figure 4. Setup of specimens in Laser Roll Welding
Table 2. Process parameters for Laser Roll Welding

Welding Processes 38
2.2. Process parameters
In this study, to investigate the effect of the welding condition on the joint properties of two
types of zinc coated steel and aluminum alloy by Laser Roll Welding, welding speed and
roll pressure were varied. Process parameters are shown in Table 2.
Pulsed laser was used by controlling a 2.4 kW CO2 laser with continues wave. Laser peak
power, duty cycles and frequency were constant 2.0 kW, 75 % and 150 Hz respectively.
These parameters have been optimized in the previous study. Beam spot shape was an
quasi-elliptical with major axis of 3.5 mm and minor axis of 2.5 mm. Distance between the
center of laser beam spot and the pressurizing axis of roller was 25 mm. Welding speed was
varied from 3.3 to 15.0 mm/s. Overlapped width was 3 mm, and laser beam was irradiated
the center of overlap. The roller is mounted with a calibrated compression spring for applying
predetermined roll pressure. The roll pressure was calculated by assuming that the contact area
between the roller and the steel sheet surface was a rectangle of 15 mm
2
. Roll pressure was
varied from 100 to 175 MPa. For the sake of the protection of condenser lens and the oxidation
prevention of Laser Roll Welded joints, argon gas with flow rate of 25 l/min was used.
2.3. Evaluation method of Laser Roll Welded joints
After welding, Laser Roll Welded specimens were cut across the lap joint seam for
macrostructure and microstructure observation. Etching with 3% nital was made for
observation of the interface layer. Hardness test and electron-probe microanalysis (EPMA)
were conducted to analyze the interface layer and to identify the IMC present.

Figure 5. Schematic diagram of measurement method for thermal cycle at weld interface
Thermal cycle was measured using a Pt-PtRh thermocouple with a diameter of 0.3 mm. As
shown in Fig.5, a cone-shaped hole of 2 mm in diameter was drilled at the center of the
overlap width in the aluminum sheet. The hole was used for placing thermocouple at the
lower surface of the steel sheet. In addition, tensile shear test and Erichsen cupping test were
carried out to investigate the mechanical properties of welded joint. The specimens with 20
mm width were used for tensile shear test and with 77 mm square for Erichsen cupping test.

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 39
3. Experimental results and discussions
3.1. Bead appearance and cross-section of Laser Roll Welded joints
Bead appearance of Laser Roll Welded specimen with the welding speed of 8.3 mm/s and
roll pressure of 150 MPa are shown in Figs.6-7. Top bead is shown in Fig.6, and bottom bead
is in Fig.7. Furthermore, cross-section of the weld bead is shown in Fig.8. In these figures, (a)
shows GI/A6000 joint and (b) shows GA/A6000, respectively.

Figure 6. Top bead appearance of GI/A6000 and GA/A6000 Laser Roll Welded joints

Figure 7. Bottom bead appearance of GI/A6000 and GA/A6000 Laser Roll Welded joints

Figure 8. Macro cross-section of GI/A6000 and GA/A6000 Laser Roll Welded joints
As shown in Fig.6, both of the bead turned black due to the effect of the graphite spray
and the laser heating. In Fig.6 (a), striped pattern is observed on the GI sheet near the
weld bead. This is the first time this striped pattern has been observed in the past study
of Laser Roll Welding, the pattern seems to be unique to this welding of galvanized
steel sheet. A GI sheet has galvanized layer, which exists as a thin layer at the surface of
steel sheet. Thus its thought that the pattern was caused by the laser heating and the
roller pressure. On the other hand, the striped pattern doesnt exist on the GA sheet as

Welding Processes 40
shown in Fig.6 (b). This is because the coated layer of a GA sheet is alloyed zinc with
iron.
As shown in Figs.7-8 (a), partial melting and spreading of molten aluminum alloy on
the GI sheet occurs in the A6000 sheet. The GI sheet was heated by laser, and the heat
transferred to A6000 sheet by the contact of GI and A6000 sheets. Hence A6000 sheet
was supposed to be melted. In contrast, molten A6000 alloy doesnt spread on the GA
sheet.
In order to compare the wettability of the GI sheet with that of the GA sheet, bonding width
was measured from the cross-section of the weld bead as shown in Fig.9 (a). Effect of the
welding speed and the roll pressure on the bonding width of GI/A6000 and GA/A6000 joints
are shown in Fig.9 (b) and (c), respectively.

Figure 9. Effect of welding speed on bonding width at different roll pressures
Its common with Fig.9 (b) and (c) that the bonding width decreased as the welding speed is
increased from 5.0 to 11.7 mm/s. This is because heat input decreases as the welding speed
increased; the meltage of A6000 sheet also decreased. On the whole, the bonding width of
GA/A6000 joints is narrower than that of GI/A6000. For example, the width with the
welding speed of 8.3 mm/s and roll pressure of 150 MPa of GI/A6000 joint is 4.6 mm, and
that of GA/A6000 is 3.5 mm. This is considered to be produced from the difference in the
wettability of the surface of the GI and GA sheet.
3.2. Microstructures at weld interface
Interface microstructures of GI/A6000 joint with welding speed of 6.7 mm/s and roll
pressure of 150 MPa are shown in Fig.10.

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 41

Figure 10. Microstructures at joint interface of GI and A6000 sheet
IMC layer was confirmed at the joint interface between the GI and the A6000 sheet. The IMC
layer at the edge of the GI sheet is shown in Fig.10 (a), at the center of the laser beam spot is
shown in (b) and at the edge of the A6000 sheet is show in (c). The IMC layer thicknesses in
each place were 4.1 m, 13.0 m, and 4.9 m respectively. Therefore, the thickness right
under the center of laser beam spot was the thickest. Due to the high energy density near the
center of the beam spot of the CO2 laser used, its thought that the area under the center of
the beam spot is heated the highest temperature and cooled the slowest speed.
Interface microstructures of GA/A6000 joint with welding speed of 8.3 mm/s and
roll pressure of 150 MPa are shown in Fig.11. IMC layer was also observed at the
interface between the GA and the A6000 sheet. In addition, the thickness right under the
center of laser beam spot was the thickest, and this was the same as that of the case of
GI/A6000 joint.

Figure 11. Microstructure at joint interface of GA and A6000 sheet

Welding Processes 42
3.3. Electron-probe microanalysis (EPMA) of interlayer
As the optical microscope doesnt reveal the details of the IMC layers, EPMA of iron,
aluminum and zinc across joint interface were made to identify IMCs and existence of zinc.
The results for GI/A6000 and GA/A6000 joints at different welding speeds with roll pressure
of 150 MPa are shown in Figs.12-13. Bottom pictures show the SEM images of the IMC layer.
In these figures, the EPMA results with the welding speed of 8.3 mm/s are shown in (a),
with the welding speed of 10.0 mm/s are shown in (b).

Figure 12. Results of EPMA at interlayer of GI/A6000 joints
From EPMA results of iron, aluminum and zinc across interface, IMCs and existence of zinc
were identified. At the center of these EPMA results which are shown by the layer A to C in
Fig.12 as an example, the signal intensity of Fe and Al has changed similarly. At the layer A,
the intensity of Fe decreases rapidly, and Al rises. At the layer B, the IMCs are observed as
stepped lines. At the layer C, the intensity of Fe decreases further and Al rises. From the
estimation of composition from stepped lines at the layer B, its suggested that main IMCs
were brittle FeAl3 and Fe2Al5.
When the welding speed is increased to 10.0 mm/s, zinc can be seen in aluminum alloy.
From this result, its thought that zinc tends to diffuse into aluminum when the welding
speed becomes more than 10.0 mm/s. This reason is considered as follows.
Most of the zinc layer is vapored by the laser heating from the faying surface of the GI sheet
at slow welding speeds. On the other hand, at fast welding speeds, the diffusion amount of

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 43
the zinc into aluminum alloy by the roll pressure is larger than the evaporation amount of
the zinc by the laser heating.

Figure 13. Results of EPMA at interlayer of GA/A6000 joints
As shown in Fig.13 (a), the zinc line has higher peak than any other zinc line at the interface
of the GA sheet side. This is because the zinc on GA sheet surface exists as Fe-Zn alloy. Then
its suggested that zinc was hard to be evaporated by the laser heating, and a lot of zinc was
remained at the interface of the GA sheet side.
3.4. Vickers hardness measurements
From the EPMA results, it was presumed that main IMCs were brittle FeAl3 and Fe2Al5.
However, in order to obtain further evidence, Vickers hardness measurement was
conducted. SEM image of GI/A6000 and GA/A6000 weld interface after the measurement
with welding speed of 8.3 mm/s and roll pressure of 150 MPa are shown in Fig.14 (a) and
(b), respectively.
Indentation size became small in order of the base metal of A6000, zinc coated steel, and the
IMC layer. In particular, the indentation size of the IMC layer is much smaller than that of
both base metals. As shown in Fig.14 (a), the Vickers hardness of the base metal of the GI,
the A6000, and the IMC layer were 137 Hv, 94 Hv and 940 Hv on average respectively. The
results show a large difference between IMC layer hardness and the surrounding base
metals. The IMC hardness was about 10 times more than the A6000 base metal and 7 times
more than the GI base metal. In the same way, as shown in Fig.14 (b), the Vickers hardness

Welding Processes 44
of the base metal of the GA, the A6000, and the IMC layer were 141 Hv, 97 Hv and 857 Hv
on average respectively, the IMC layer was the hardest. Because the hardness of these IMC
layers are between 892 Hv of FeAl3 and 1013 Hv of Fe2Al5 according to Table 1, its thought
that FeAl3 and Fe2Al5 are mainly formed at the interface as above.

Figure 14. Results of Vickers hardness measurement at joint interface
3.5. Effect of welding speed on thickness of intermetallic compound layer
Effect of the welding speed on the IMC layer thickness of GI/A6000 and GA/A6000 joints at
different roll pressure are shown in Fig.15 (a) and (b), respectively. The thicknesses were
measured right under the center of the laser beam spot.

Figure 15. Effect of welding speed on IMC layer thickness of Laser Roll Welded joints interface
The IMC layer thickness decreases significantly as the welding speed is increased from 5.0 to
11.7 mm/s regardless of the roll pressure as shown in Fig.15 (a) and (b). This is because heat
input decreases as the welding speed in increased. Therefore, this result indicates that the
IMC layer thickness could be suppressed by heat input as was mentioned in previous study
of Laser Roll Welding of steel to aluminum alloy.
3.6. Effect of roll pressure on thickness of intermetallic compound layer
Effect of the roll pressure on the IMC layer thickness of GI/A6000 and GA/A6000 joints at
constant welding speed of 8.3 mm/s is shown in Fig.16.

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 45
The IMC layer thickness of both joints increases as the roll pressure is increased from 100 to
150 MPa. This is because increment of the roll pressure augments the contact between the
zinc coated steel and A6000 sheet in this region. In contrast, the IMC layer thickness decreases
as the roll pressure is increased from 150 to 175 MPa. This reason is considered as follows.

Figure 16. Effect of roll pressure on IMC layer thickness of Laser Roll Welded joints interface
The zinc coated steel and the A6000 sheet contact most widely at roll pressure of 150 MPa.
However, when the roll pressure becomes more than 150 MPa, the steel sheet might be
curved greatly by high roll pressure. Thus, its thought that the contact area between the
zinc coated steel and the A6000 sheet is narrowed in this region.
3.7. Effect of types of zinc coated steel on thickness of intermetallic
compound layer
As shown in Fig.16, the IMC layer thickness of GA/A6000 joints is thicker than that of
GI/A6000. This difference is attributed to the evaporation of zinc on the surface of each zinc
coated steel sheets.
The boiling point of aluminum, iron and zinc are 2477
o
C, 2887
o
C and 906
o
C respectively, that
of zinc is far below those of aluminum and iron. Hence, the zinc on the surface of zinc coated
steel sheet is melting, heat of fusion of 7.12 kJ/mol, and evaporating, heat of vaporization of
113.4 kJ/mol (The Japan Institute of Metals, 2004), in the process of Laser Roll Welding.
However, because of the zinc on the surface of the GA sheet exists as Fe-Zn alloy, its hard to
be evaporated by the laser heating. Thus, laser energy is little-used for the zinc evaporation,
and heat is conducted to A6000 sheet. On the other hand, the zinc on the surface of the GI
sheet is easier to be evaporated than that of the GA. Therefore, its thought that more laser
energy for the zinc evaporation is used, and the heat conduction to A6000 sheet decreases.
3.8. Discussion of formation of intermetallic compound layer by thermal cycle at
weld interface
Thermal cycle at the weld interface was measured to discuss about the effects of the welding
speed and the roll pressure on the IMC layer thickness of Laser Roll Welded joints. The

Welding Processes 46
results of GI/A6000 and GA/A6000 joints at constant roll pressure of 150 MPa is shown in
Fig.17 (a) and (b), respectively.
When the welding speed increases from 5.0 to 11.7 mm/s, different thermal cycles were
obtained as shown in Fig.17 (a) and (b). Then, in order to quantitative the shape of these
thermal cycles, peak temperature and holding time more than 500
o
C were focused on.

Figure 17. Effect of welding speed on interface thermal cycles of Laser Roll Welded joints

Figure 18. Effect of welding speed on peak temperature and holding time more than 500
o
C at interface
of GI/A6000 joints
The peak temperature and the holding time more than 500
o
C of the thermal cycles in Laser
Roll Welding of the GI and the A6000 at different welding speeds are shown in Fig.18. When
the welding speed is increased from 5.0 to 11.7 mm/s, the peak temperature decreases from
850 to 680
o
C, and the holding time more than 500
o
C shortens from 2.8 to 1.1 sec at the weld
interface. Hence the reduction of the IMC layer thickness by the increment of welding speed
was attributed to the decline of the peak temperature and the shortening of the holding
time. The similar results were obtained in Laser Roll Welding of the GA and the A6000.
When the welding speed is slow, there is excessive heat input and the cooling rate is slow.
This provides surplus time for the formation of a thick interlayer containing a large amount
of the Al-rich brittle IMCs. When the welding speed is fast, there is suitable heat and time

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 47
for melting of the aluminum and the diffusion process to take place. Therefore, its thought
that the change of the thermal cycle at the interface affect the formation of the IMC layer
when the welding speed is varied.
In addition, when Fig.17 (a) is compared with (b), the peak temperatures of GA/A6000 joints
are higher than those of GI/A6000. This fact caused the formation of thicker IMC layer in
GA/A6000 joints than in GI/A6000 as shown in Fig.16.
Effect of the roll pressure on the interface thermal cycles of GA/A6000 joints at constant
welding speed of 8.3 mm/s is shown in Fig.19. Thermal cycles are shown in Fig.19 (a) and
the peak temperature and the holding time more than 500
o
C of them are in (b). The reason
of the changing of IMC layer thickness by the increment of the roll pressure as shown in
Fig.16 is considered by using Fig.19.

Figure 19. Effect of roll pressure on interface thermal cycles of GA/A6000 joints
From 100 to 150 MPa, the peak temperature and the holding time increase as the roll
pressure is increased in Fig.19. Hence the IMC layer thickness increased as the roll pressure
is increased in Fig.16. This is because increment of the roll pressure augments the contact
between zinc coated steel and A6000 sheet in this region as above. In contrast, from 150 to
175 MPa, the peak temperature rises, but holding time shortens as the roll pressure is
increased in Fig.19. Its supposed that this tendency caused the decrement of IMC layer
thickness as the roll pressure is increased at high region in Fig.16.
3.9. Results of tensile shear test
Tensile shear test was conducted to investigate the influence of the welding conditions on
the weldability. The tensile shear specimens were prepared by cutting the welded specimen
with 20 mm width.
Fig.20 shows the tensile shear specimen after testing of GI/A6000 joints. Failure in base
metal of the GI sheet is shown in Fig.20 (a), and failure in interface is shown in (b). Its

Welding Processes 48
found that the base-metal-failure specimen failed far from the weld bead as shown in
Fig.20 (a).

Figure 20. Tensile shear specimen after testing of GI/A6000 Laser Roll Welded joints
Results of tensile shear test of GI/A6000 and GA/A6000 joints at various welding speed with the
roll pressure of 150 MPa is shown in Fig.21 (a) and (b), respectively. Here the tensile shear
strength was converted into the tensile shear load per millimeter of weld length, N/mm, as
adopted by Peyre et al. (2007) and Sasabe et al. (2007). Failure in the base metal of the zinc
coated steel sheet is shown as a circle in the figure, and failure in interface is shown as a solid
mark.

Figure 21. Effect of welding speed on tensile shear strength for Laser Roll Welded joints
When the welding speed is between 6.7 and 10.0 mm/s, the tensile shear strength shows
high, and there are many specimens failed in the base metal. This result indicates that the
IMC layer thickness is appropriate at these welding speeds. As shown in Fig.21 (a),
specimens failed in the base metal could be obtained from 6.7 to 10.0 mm/s. In this
region, the IMC layer thickness was less than 10 m, and failure of specimen occurred in
the base metal of the steel sheet. Therefore, its thought that the IMC layer thickness
should be less than 10 m to get a good joint. As shown in Fig.21 (b), specimens failed in
base metal were confirmed when the IMC layer thickness was less than 10 m. This
result corresponded to the articles by other researchers (Bruckner, 2005; Furukawa,
2005).
Additionally, when Fig.21 (a) is compared with (b), the tensile shear strength of GA/A6000 joints
is lower than those of GI/A6000. There are three points which can be considered as this reason.

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 49
The first is the bonding width of GA/A6000 joints was narrower than that of GI/A6000 as
shown in Fig.9. Therefore, when tensile shear load was applied to the joints with narrow
bonding width, the load concentrated to the weld. The second is the IMC layer thickness of
GA/A6000 joints was thicker than that of GI/A6000 as shown in Fig.16. Moreover, the IMC
layers were mainly composed of Al-rich brittle IMCs from the results of the EPMA and the
Vickers hardness. The third is the residual zinc at the interface of GA/A6000 joints
influenced the tensile shear strength. From the results of the EPMA, a lot of zinc was
remained as Fe-Zn alloy layers at the interface of GA sheet side. The zinc coated layer of the
GA sheet is formed from the steel side in order of -phase (Fe3Zn10), 1- (Fe5Zn21), - (FeZn7)
and - (FeZn13) (The Iron and Steel Institute of Japan, 1982). Since each phase has a difference
in generation speed, the -phase is thin and alloy layers are composed mostly of - and -.
Their phases are hard and brittle. Therefore, the interface of the GA/A6000 joints is weaker
than that of the GI/A6000.
The maximum tensile shear strength of GI/A6000 joint, 162 N/mm, was obtained at the
welding speed of 8.3 mm/s and the roll pressure of 150 MPa. This strength is equal to the
tensile strength of the GI sheet, 157 N/mm, and 63% of the A6000 sheet, 256 N/mm. On
the other hand, the maximum strength of GA/A6000 joint, 160 N/mm, was obtained at
the welding speed of 10.0 mm/s and the roll pressure of 100 MPa. This strength is equal
to the tensile strength of the GA sheet, 164 N/mm, and 62% of the A6000 sheet, 256
N/mm.

Figure 22. Effect of roll pressure on tensile shear strength for GI/A6000 joints
In addition, results of tensile shear test at various roll pressure with welding speed of 8.3
mm/s is shown in Fig.22. All specimens were failure in the base metal of the GI sheet except
for the roll pressure 100 MPa. Its considered that the IMC layer thickness is less than 10 m
at these roll pressures as shown in Fig.16. Therefore, it seems that there is little influence of
the roll pressure on the tensile shear strength.

Welding Processes 50
3.10. Discussion of relation between intermetallic compound layer thickness
and tensile shear strength
From the above results, it has became clear that there is a close relation to the IMC layer
thickness to the tensile shear strength also in Laser Roll Welding of zinc coated steel and
aluminum alloy. There, the relationship between the IMC thickness and the tensile strength
was discussed about all joints in this experiment. Effect of the IMC layer thickness on the
tensile shear strength of GI/A6000 and GA/A6000 joints with different welding speed and
roll pressure is shown in Fig.23 (a) and (b), respectively.
As above, only when the IMC layer thickness was less than 10 m, the base-metal-
failure specimens were obtained. When the IMC layer thickness is from 4 to 6 m, high
tensile shear strength could be obtained. When the IMC layer is thicker than 6 m, the
tensile strength is decline. This is because the increment of the brittle IMCs at the joint
interface might lead to weaken of the welded joint, and the joint isnt able to resist a
heavy load.
On the other hand, when the IMC layer is thinner than 4 m, the tensile strength also
declines. In this case, due to low heat input might lead to incomplete welding at the joint
interface. As shown in Fig.9, the bonding width is decreased by the decrement of the heat
input.

Figure 23. Effect of IMC layer thickness on tensile shear strength for Laser Roll Welded joints
3.11. Results of Erichsen cupping test
Finally, in order to investigate the formability of GI/A6000 and GA/A6000 joints, Erichsen
cupping tests were carried out. The specimens were prepared by cutting the welded joints
with the welding speed of 8.3 mm/s and roll pressure of 150 MPa into 77 mm square. A
punch is pushed into the specimen. When the specimen is failed somewhere, the cupping
height is evaluated as Erichsen value.

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 51

Figure 24. Erichsen cupping test specimen after testing of GI/A6000 joint
The Erichsen cupping test specimen after testing of GI/A6000 joint is shown in Fig.24. The
specimen was failed at HAZ of the A6000 sheet side and the Erichsen value was 7.9 mm.
With the GI base metal, the Erichsen value was 11.9 mm; with the A6000 base metal, the
value was 8.6 mm. Therefore, this value was 92% of the base metal of A6000 sheet. In
contrast, the specimen of GA/A6000 joint was failed at interface and the value was 3.6 mm.
The same tendency was seen in the tensile shear test.
4. Conclusions
The present study is focused on the dissimilar metal joining of zinc coated steel and
aluminum alloy by Laser Roll Welding. The following conclusions can be drawn.
1. The IMC layer was observed at the interface of all welded joins. It was suggested that
most of the IMC's are brittle FeAl3 and Fe2Al5 from the results of EPMA and Vickers
hardness measurement. As the welding speed was faster than 10.0 mm/s, zinc was
confirmed in aluminum alloy.
2. Increase in the welding speed led to decrease the bonding width and the IMC layer
thickness at the joint interface. When the roll pressure was increased, the IMC thickness
at the pressure of 150 MPa was the thickest. The IMC layer thickness of GA/A6000 joints
was thicker than that of GI/A6000 on the whole.
3. Increase in the welding speed led to lowering of the peak temperature and shortening
of the holding time more than 500
o
C at the interface. The peak temperature at the roll
pressure of 175 MPa was the highest, and the holding time at the pressure 150 MPa was
the longest. The peak temperature of GA/A6000 joints was higher than that of GI/A6000
at the same welding condition.
4. When the IMC layer was less than 10 m, failure of specimen occurred at the base metal
of zinc coated steel in tensile shear test. The joint properties of GI/A6000 joints were
better than those of GA/A6000 from the results of tensile shear test and Erichsen
cupping test.

Welding Processes 52
5. The welding speed influenced the joint performance such as the IMC layer thickness
and the tensile shear strength to a greater degree than the roll pressure in Laser Roll
Welding.
Author details
Hitoshi Ozaki
Graduate School of Engineering, Mie University, Japan
Muneharu Kutsuna
Advanced Laser Technology Research Center Co., Ltd., Japan
Acknowledgement
The authors would like to thank Mr. S. Nakagawa and Mr. K. Miyamoto from Research
Center, Nissan Motor Co., Ltd. for their support in this research.
5. References
Bruckner, J. (2005). Cold Metal Transfer Has a Future Joining Steel to Aluminum. Welding
Journal, Vol.84, No.6, pp.38-40, ISSN 0043-2296
Dharmendra, C., Rao, K. P., Wilden, J. & Reich, S. (2011). Study on Laser Welding-brazing of
zinc Coated Steel to Aluminum Alloy with a Zinc Based Filler. Material Science and
Engineering A, Vol.528, pp.1497-1503, ISSN 0921-5093
Fan, J., Thomy, C. & Vollertsen, F. (2011). Effect of Thermal Cycle of the Formation of
Intermetallic Compounds in Laser Welding of Aluminum-Steel Overlap Joints. Physics
Procedia, Vol.12, pp.134-141, ISSN 1875-3892
Furukawa, K. (2005). Welding Process of Iron-Aluminum. Welding Technology, Vol.53, No.8,
pp.94-102, ISSN 0387-0197
Katayama, S. (2004). Laser Welding of Aluminum Alloys and Dissimilar Metals. Journal of
Light Metal Welding & Construction, Vol.42, No.1, pp.16-25, ISSN 0368-5306
Katoh, K. & Tokisue, H. (2004). Dissimilar Friction Welding of Aluminum Alloys to Other
Materials. Journal of Light Metal Welding & Construction, Vol.42, No.2, pp.11-18, ISSN
0368-5306
Massalski, T. B. (1986). Binary Alloy Phase Diagrams Volume 1, American Society for Metals,
p.148, ISBN 0871702614, Ohio
Nishimoto, K., Atagi, K., Fujii, H. & Katayama., S. (2005). Laser Pressure Welding of
Dissimilar Metals Welding. Proceedings of the 63rd Laser Materials Processing Conference,
pp.133-138, ISBN 4947684585, May, 2005
Ohashi, O. (2004). Diffusion Bonding of Aluminum to Different Metals. Journal of Light Metal
Welding & Construction, Vol.42, No.2, pp.19-23, ISSN 0368-5306

Dissimilar Metal Joining of Zinc Coated Steel and Aluminum Alloy by Laser Roll Welding 53
Okamura, H. & Aota, K. (2004). Joining of Dissimilar Materials with Friction Stir Welding,
Journal of Light Metal Welding & Construction, Vol.42, No.2, pp.1-10, ISSN 0368-
5306
Okita, T. (2004). Resistance Welding of Aluminum Alloys to Dissimilar Metals. Journal of
Light Metal Welding & Construction, Vol.42, No.1, pp.2-15, ISSN 0368-5306
Ozaki, H., Hayashi S. & Kutsuna M. (2008). Laser Roll Welding of Dissimilar Metal Joint of
Titanium to Aluminum Alloy. Quarterly Journal of Japan Welding Society, Vol.26, No.1,
pp.24-30, ISSN 0288-4771
Ozaki, H., Ichioka R. & Kutsuna M. (2007). Laser Roll Welding of Dissimilar Metal Joint of
Low Carbon Steel and Titanium. Quarterly Journal of Japan Welding Society, Vol.25, No.1,
pp.173-178, ISSN 0288-4771
Ozaki, H. & Kutsuna M. (2007). Laser Roll Welding of Dissimilar Metal Joint of Low Carbon
Steel to Aluminum Alloy Using 2kW Fiber Laser. Quarterly Journal of Japan Welding
Society, Vol.25, No.4, pp.473-479, ISSN 0288-4771
Peyre, P., Sierra, G., Deschaux-Beaume, F., Stuart, D. & Fras, G. (2007). Generation of
Aluminum-steel Joints with Laser-induced Reactive Wetting. Material Science and
Engineering A, Vol.444, pp.327-338, ISSN 0921-5093
Rathod, M. & Kutsuna M. (2003). Laser Roll Bonding of A5052 Aluminum Alloy and SPCC
Steel. Quarterly Journal of the Japan Welding Society, Vol.21, No.2, pp.282-294, ISSN 0288-
4771
Rathod, M. J. & Kutsuna M. (2004). Joining of Aluminum Alloy 5052 and Low-Carbon
Steel by Laser Roll Welding. Welding Journal, Vol.83, No.1, pp.16s-26s, ISSN 0043-
2296
Sasabe, S., Iwase, T., Matsumoto, T., Hattori, Y. & Miono, T. (2007). Dissimilar Metal Joining
of Aluminum Alloys to Steel by using Newly-developed Hot-dip Aluminized Steel
Sheet. Journal of Light Metal Welding & Construction, Vol.45, No.2, pp.23-33, ISSN 0368-
5306
Satou, D. (2004). Aluminum Clad of Explosive Welding. Journal of Light Metal Welding &
Construction. Vol.42, No.1, pp.26-30, ISSN 0368-5306
The Iron and Steel Institute of Japan (1982). Iron and Steel Handbook Vol. 6 (3
rd
Edition),
Maruzen, pp.421-434, Tokyo
The Japan Institute of Metals (2004). Metals Data Book (Revised 4
th
Edition), Maruzen, p.11,
ISBN 4621073672, Tokyo
Torkamany, M., Tahamtan, S. & Sabbaghzadeh, J. (2010). Dissimilar Welding of Carbon Steel
to 5754 Aluminum Alloy by Nd:YAG Pulsed Laser. Materials and Design, Vol.31, pp.458-
465, ISSN 0261-3069
Yamamoto, S. (2012). Welding in the World Picking up from Foreign Magazine. Welding
Technology, Vol.60, No.3, (2012), p. 75, ISSN 0387-0197
Yan, S., Hong, Z., Watanabe, T. & Jingguo, T. (2010). CW/PW Dual-beam YAG Laser
Welding of Steel/Aluminum Alloy Sheets. Optics and Lasers in Engineering, Vol.48,
pp.732-736, ISSN 0143-8166

Welding Processes 54
Yasuyama, M., Ogawa, K. & Taka, T. (1996). Spot welding of aluminum and steel sheet with
insert of aluminum clad steel sheet Part 1. Quarterly Journal of Japan Welding Society,
Vol.14, No.2, pp.314-320, ISSN 0288-4771
Chapter 3




2012 Guo, licensee InTech. This is an open access chapter distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
In situ Reaction During Pulsed Nd:YAG
Laser Welding SiC
p
/A356
with Ti as Filler Metal
Kelvii Wei Guo
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/46087
1. Introduction
As aluminum matrix composites are generally low-cost and exhibit higher specific strength,
high wearability, and good design performance property and functionality. They are widely
applied in aerospace-flight, aviation structure, and automobile and in the heat resistant-
wearable parts of engine [1-4]. Hence, a great deal of contemporary research effort is
focusing upon their development and applications, typically on the discontinuously
reinforced aluminum matrix composites like matrix with particle, short fiber, whisker and
so forth. Additionally, a great deal of attention has also been drawn into the investigation of
their secondary processing technologies like machining, joining and plastic forging. Welding
is an important process for joining these materials. There has extensive effort to be devoted
to developing appropriate process for joining the similar or dissimilar composites in
literatures [2-4]. These processes can be mainly categorized as: (i) fusion welding like arc
welding, shielding gas welding, laser welding and electron beam welding, etc.; and (ii)
solid-state welding like soldering, explosive welding, friction welding and diffusion
welding, etc. However, there still exist many problems in joining of the discontinuously
reinforced aluminum matrix composites using conventional arc-welding processes and
those high energy density welding methods like: laser welding and electron beam (EB)
welding. These problems include: (i) the formation of poor weldment and the unsatisfactory
properties of welded joints mainly due to the high viscosity and poor flowability of the
liquid welding pool causing mixing difficulty of the composite base material with filler
materials; (ii) the occurrence of micro-segregation or inhomogeneous distribution of the
reinforcement phases of SiCp, Al2O3p, AlN and etc., and whiskers like SiCw typically owing
to the rejection by their solidification front in the welding poor as cooling down, which
subsequently prompts for many micro and macro defects in the weld and very poor

Welding Processes 56
properties of the welded joints; (iii) the formation of aluminum carbide mainly as a result
of harmful interfacial reaction between aluminum matrix and reinforcement phases; and (iv)
so on. A typical example of the interfacial reaction likely to have pernicious effects on the
mechanical and chemical behavior of the composite is 4Al + 3SiC Al4C3 + 3Si. This is
owing to: (a) the formation of brittle and weak aluminum carbide Al4C3 in the interfacial
reaction sacrifices the reinforcement-materials in the composite; and (b) the unstable
aluminum carbide in wet environments causes corrosion of the composite because of its
rapid hydrolysis etc.
Aiming at developing or improving the conventional welding technique, this paper studies
the technique of welding the stir-cast aluminum matrix composite SiCp/A356 by Nd:YAG
laser welding with pure titanium as filler. This study has been specifically concerned on the
in situ reinforcement effect of Ti on the microstructures of laser welded joints, which have
been analyzed by means of Scanning Electron Microscope (SEM), Transmission Electron
Microscope (TEM) and Electron Diffusion X-ray analysis (EDX) etc. The study aims at
providing some ground works for further studies in this field.
2. Experimental material and process
2.1. Experimental material
Stir-cast SiCp/A356 aluminum matrix composite, reinforced with 20 % volume fraction SiC
particle of 12 m mean size, was used as the welding specimens. The tensile strength of
such specimens was 240 MPa and their solid-liquid phase transformation temperature was
in the range of 562.6578.3 C. Figure 1 showed their corresponding microstructure while
Table 1 tabulated the chemical composition of the A356 matrix alloy. Pure titanium was
used as the filler metal.

Figure 1. Microstructure of SiCp/A356 aluminum matrix composite

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 57
Composition (Wt %)
Si Mg Ti Al
6.5~7.5 0.3~0.5 0.08~0.2 Bal.
Table 1. Composition of A356
2.2. Experimental process
The stir-cast aluminum matrix composite specimens were individually wire-cut to the size
of 3 mm 10 mm 35 mm (Fig. 2). The quench-hardened layer induced by wire-cut and the
oxide on the surfaces of specimens were polished away by 400 # emery cloth. The pure
titanium filler was then machined to depth 3 mm width 10 mm thickness of 0.15 mm, 0.3
mm, 0.45 mm, 0.5 mm, 0.6 mm and 0.75 mm, respectively The specimens were then
mounted into a clamping devise on the platform of a GSI Lumonics Model JK702H Nd:YAG
TEM00 mode laser system, and their welding surfaces were properly cleaned by acetone and
pure ethyl alcohol so as to remove any possible contaminant. The prepared pure titanium
filler was also thoroughly cleaned and carefully sandwiched between the two composite
specimens in the clamp. Thereafter, specimens were welded immediately by the Nd:YAG
laser with wavelength of 1.06 m, defocused distance of 10 mm so as to give a focus spot
diameter of approximately 1.26 mm, laser fluence energy 2 J, frequency 25 Hz, and pulse
duration 4 ms. In the welding, the relative moving speed of the laser and the welding pieces
(i.e. feedrate) was set at 300 mm/min

Figure 2. Schematic illustration of laser welding with Ti filler
Tensile strength of the joint was performed by a MTS Alliance RT/30 electron-mechanical
material testing machine with a straining velocity of 0.5mm/min.The cross-section of
welded joints was wire-cut for optical microscope investigation, and Scanning Electron
Microscope (SEM) and Transmission Electron Microscope (TEM) analyses. SEM was used
to analyze in detail the microstructure at the weld joints and the fractured tensile test-
pieces of the joints. Optical microscope was used for observing the structure of a large
area. TEM and Electron Diffusion X-ray analysis (EDX) were used to analyze the interface
between the newly-formed phases and aluminum matrix, the distribution of chemical
elements and spectra at the joints. The Nd:YAG laser with the similar setting conditions
and feedrate was also used to weld the aluminum matrix composite specimens without
any filler.

Welding Processes 58
3. Results and discussion
3.1. Microstructures and properties of welded joints
The microstructure of the in situ reinforcement of Ti by Nd:YAG laser welding with 0.3 mm
thick Ti filler was shown in Fig. 3. Appearance of in situ reinforcement particles distributed
uniformly in the welded joint was seen. It also showed the disappearing of the drawbacks
like incomplete fusion and pernicious phase Al4C3. These subsequently resulted in higher
tensile strength (Table 2) of the joint. Comparatively, the reinforcement particles distributed
more compactly than that of parent composite (cf. Fig. 1 and Fig. 3). The relatively more
highly compacting reinforcement particles improved distinctly the properties of welded
joints. The presence of Ti effectively improved the flowability of the liquid welding pool and
the newly formed in situ reinforcement particles (Fig. 3) replaced those initial reinforcement
particles (Fig. 1). Those dimples appeared in the SEM of the corresponding fractured surface
(Fig. 4) suggested that: (i) the new-formed reinforcement particles had been perfectly wet;
and (ii) the harmful composite structure of the initial welding viz.
reinforcement/Ti/reinforcement had been changed to the state of reinforcement/matrix
/reinforcement. XRD pattern of the fractured surface (Fig. 5) of the weld joint did not give
sign of any harmful phase or brittle phase of Al4C3. This suggested the occurrence of
effective interface transfers between reinforcement particles and matrix in the laser welded
joint that subsequently provided favorable welding strength (Table 2). By the newly formed
in situ reinforcement particles as detected by EDX (Table 3) and the intensity spectra shown
in Fig. 5, the newly-formed reinforcement particle in the weld was identified as TiC.

Figure 3. Microstructure of in situ reinforcement by laser welding with 0.3 mm thick Ti filler
Thickness of Ti filler (mm) 0.00 0.15 0.3 0.45 0.60 0.75
Mean strength (MPa) 76-91 126-135 168-180 143-160 107-113 79-96
Table 2. Mean tensile strength of laser welded specimens with various Ti filler thicknesses

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 59

Figure 4. Fractograph of the laser welded joint with 0.3 mm thick Ti filler

Figure 5. XRD pattern of the fracture surface for laser welding with 0.3 mm thick Ti filler
Element Ti Si Al
Wt (%) 81.84 4.42 Bal.
Table 3. EDX analysis of newly-formed particle in the laser weld with 0.3 mm thick Ti filler

Figure 6. Macro-structure of the laser welded joint with 0.3 mm thick Ti filler
3mm
A
B
B
C
C D D

Welding Processes 60

Figure 7. Microstructures of the different areas in the laser weld with 0.3 mm thick Ti filler
(a) Area A (b) Area B (c) Area C

Figure 8. Microstructure of the weld with no-filler
Figure 6 illustrates the macro-structure of welded joint with Ti filler. Basically, the weld
consisted of three main areas, namely: the in situ reinforcement area A, the two transitional
areas B and C, and the no-in situ reinforcement area D. Their individual microstructures
were shown in Fig. 7. The microstructures indicated the initial reinforcement SiC particles
were completely replaced by the newly-formed in situ reinforcement TiC particles that
mainly resulted in the formation of the in situ reaction area A (Fig. 7a). In the area B, the
newly-formed in situ reinforcement TiC particles and the initial reinforcement SiC particles
were coexistent (Fig. 7b). In the area C, very little newly-formed in situ reinforcement TiC
particles were found (Fig. 7c). In the area D, there only existed the initial reinforcement
particles SiC (Fig. 1). It was found that the typical pernicious acicular Al4C3 microstructure
had been effectively alleviated in the welded area. Hence, it improved markedly the
properties of the welded joints and their achievable tensile strength was up to 180 MPa
(Table 2) that was about 75 % of the strength of SiCp/A356. The microstructure (Fig. 8) of the
traditional Nd:YAG laser welding with no-filler showed that there were lots of pernicious
acicular Al4C3 in the weld, which led to lower weld joint tensile strength (Table 2), i.e. only
as high as 91 MPa (that was about 37.9 % of its parent SiCp/A356).
(a) (b) (c)

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 61
3.2. Element distribution in the transition area
Figure 9 illustrates the element distribution of the area B in the weld as shown in Fig. 6 and
Fig. 7b. It showed the newly-formed in situ reinforcement particles were around the initial
reinforcement SiC particles which offered a high density nucleus area for the nucleation of
in situ TiC. The stirring effect in the welding pool by laser irradiation and the initial
reinforcement SiC particles would be replaced by the newly-formed in situ TiC (cf. Figs 9b
and 9c) following the chemical reaction process of:
Ti ( l ) + SiC ( s ) TiC ( s ) + Si ( s )

Figure 9. Element distribution of B area in the weld; (a) micrograph of the area B (b) Ti element face
distribution (c) Si element surface distribution
(a)
(b) (c)

Welding Processes 62
According to the free energy for forming the SiC, TiC and Al4C3 carbides as elaborated in
Fig. 10 of literatures [5, 6, 7], the free energy required to form TiC was much lower than that
for Al4C3 when the reaction temperature was above 800 C. The affinity between Ti and C in
the Nd:YAG laser welding was therefore greater than that of Al and C. The chemical
reaction between Ti and SiC in the welding pool would subsequently take precedence over
the reaction between Al and SiC and thus resulted in restraining the formation of the
pernicious acicular Al4C3.

Figure 10. Free energy of formation of several metallic carbides [5, 6, 7]
3.3. Influence of Ti filler thickness
The microstructures of in situ reinforcement with various thicknesses of Ti filler were
illustrated in Fig. 11 and its corresponding fractographs were shown in Fig. 12. It
illustrated that the amount of the formed in situ TiC was distinctly increased with the
increase in the thickness of Ti filler. Test indicated that maximum strength of welded
joints (Table 2) was achieved at Ti filler thickness 0.3 mm (Figs. 3 and 4). This was because
the newly-formed in situ reinforcement particles TiC were uniformly distributed in the
weld and the initial irregular (mostly in hexagonal shape as shown in Fig. 1)
reinforcement SiC particles in the weld were no longer observed (Fig. 3). Moreover, the
pernicious acicular Al4C3 was successfully restrained (Figs 3 and 7a). At the thickness of Ti
filler below 0.3 mm, there was little sign of the newly-formed in situ reinforcement TiC
particles to be observed (Fig. 11a) and a number of pernicious acicular Al4C3 were formed
in the weld. When the thickness of Ti filler was just beyond 0.3 mm, the properties of the
joints tended to become poorer again (Fig. 12b). This was because the input energy was
mainly used for melting the Ti filler and resulted in the coexistence of coarse columnar
crystals and fine equiaxed crystals (Fig. 11b). When the thickness of Ti filler was further
increased (Fig. 11c), the percentage of liquid Ti in the welding pool would also be
increased. Subsequently, the weld zone would form coarser columnar crystals, as
displayed in the SEM micrograph of Fig. 11c, after the resolidification of the melt. From
the phase diagram (Fig. 13) of Ti-Al binary system [8], it can be anticipated that increasing
the content of Ti would lead to the intermetallic compounds like TiAl and Ti3Al, etc., to be

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 63
formed during the Nd:YAG laser welding. As illustrated by the XRD pattern of the
fractured surface of a laser weld joint with the thicker Ti filler in Fig. 14, there were some
distributing brittle intermetallic compounds like TiAl and Ti3Al to be detected. Available
literature [9] has demonstrated that TiAl and Ti3Al are the harmful intermetallic
compounds in the weld and tend to decrease obviously the strength of welded joints.
Such harmful effect may follow the chemical reaction of: 5Ti[ Al l ] +3Al[ l ] + SiC[ s ]
TiC[ s ] + Si[ Al [l] ] + Al[ l ] + ( TiAl + Ti3Al ). Hence, too thick of the Ti filler led to: (i)
the appearance of the large block of columnar crystals in the microstructure (Fig. 15); and
(ii) the new-formed reinforcement TiC to be replaced by the melted/re-solidified Ti and
subsequently only the melted/re-solidified Ti existing in the weld. Results (Figs 3, 7 and
11) indicated that there existed an optimal thickness of Ti filler in the individually set
parameters in the Nd:YAG laser welding of SiCp/A356. With the optimal thickness of Ti
filler, the initial reinforcement particles SiC distributed in the aluminum matrix composite
SiCp/A356 would offer a highly dense nucleus area for the in situ TiC nucleation. This
would effectively alleviate the forming of intermetallic compounds like TiAl and Ti3Al in
the weld. It was in this manner that ultimately created favorable condition to provide
relatively superior strength of the welded joint to that of conventional laser welding.



Figure 11. Microstructures in the A area with various thicknesses of Ti filler; (a) =0.15 mm (b) =0.45
mm (c) =0.60 mm

Figure 12. Fractographs of welded joints with various thicknesses of Ti filler; (a) =0.15 mm (b) =0.45
mm (c) =0.60 mm
(a) (b) (c)
(a) (b) (c)

Welding Processes 64

Figure 13. Binary phase diagram of TiAl [8]

Figure 14. XRD pattern of fracture surface (=0.6 mm)

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 65

Figure 15. Columnar crystals in the laser weld with 0.6 mm thick Ti filler
3.4. TEM of the interface between in situ reinforcement particle TiC and matrix
The interface between in situ reinforcement particle TiC and the matrix was analyzed by the
TEM micrograph displayed in Fig. 16. It demonstrated a clear interface between the newly-
formed in situ reinforcement particle TiC and matrix. This clearly distinct interface (Fig. 16)
suggested the occurrence of prominent in situ reaction to integrate the reinforcement particle
with matrix (cf. Figs 4 and 16), and the high probability of successfully transferring load
from the matrix to TiC and vice versa. It also gives indication that the aluminum matrix
composite SiCp/A356 would be welded satisfactorily by Nd:YAG laser.

Figure 16. TEM of interface between in situ TiC reinforcement and the matrix for laser welding with 0.3
mm Ti filler
1m
TiC

Welding Processes 66
4. Microstructure evolution during the welding
4.1. Temperature field of laser welding
The heat is assumed to be released instantaneously at time t=0 on the surface of the
substrate. This causes a temperature rise in the material as follows [10-12]:

( )
3
2
2
0
exp
4
4
Q R
T T
t
c t
o
to
| |

= |
|
\ .
(1)
where is the material density, C is specific heat, is thermal diffusivity, is thermal
conductivity, Q is the input energy.
When temperature distribution is quasi-steady state:
( )
exp
2 2
o
o
q v
T T R x
c R t o o
| |
= +
|
\ .
(2)
During the Nd:YAG laser welding, q0 can be expressed as
( ) ( ) ( ) ( )
0
2 2
4
, ,
4
P P
q N f v N f v
D D PD
PD
q q
t t
= =

(3)
Where
( ) ( )
, N f v q is the coefficient of laser welding input energy, which is direct proportional to
number of overlaps or pulse frequency. With number of overlaps increasing, or pulse
frequency increasing with constant the feed-rate,
( ) ( )
, N f v q will be increased
synchronously. With the velocity (feedrate) increasing, number of overlaps with the
constant pulse frequency will be decreased correspondingly led to lower heat input,
( ) ( )
, N f v q will be decreased accordingly.
Eq. (2) can be rewritten as:

( ) ( )
( )
2 2
2 ,
exp
2
o
N f v P
v
T T R x
D c R PD
q
o
t o

| |
= +
|
\ .
(4)
Define
1
2 c

t o
=
, then Eq. (4) can be written as:

( ) ( )
( )
2
,
exp
2
4
o
N f v P
v
T T R x
D
PD R
q
o
t

| |
= +
|
\ .

(5)

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 67

2
Define ,
4
P
q
D
PD
t
=

(6)
then Eq. (6) can be written as

( ) ( )
( )
,
exp
2
o
N f v q
v
T T R x
R
q
o

| |
= +
|
\ .
(7)
4.2. Simulation model
4.2.1. Equations for temperature distribution
Using energy balance, a differential equation can be obtained for the steady temperature
distribution in a homogeneous isotopic medium, that is

B
x y z
K K K q
x x y y z z
u u u
| |
| | | | c c c c c c
+ + =
|
| |
c c c c c c
\ . \ .
\ .
(8)
Where the boundary conditions are
1
s e
u u = ,
2
s
s s
K q
x
u c
=
c
.
For

2
2
2 2
s s s
x y z V V s
V V s
K K K d q d q d
x y z
u u u
t u u

| |
| | | | c c c
= + +
| ` | |
c c c
\ . \ .
\ .
)
} } }
(9)
After Eq. 9 is discrete for the element, and according to
1
0
n
e
e
ot o
t
=
= =
, it will be obtained

( ) ( )
s s K c r
C
K K K
s B c r
u u
u u u u u u
= + + +
(10)
where S: isothermal boundary, B: the heat-input, c: the conductive and r: the irradiative.
4.2.2. Hypothesis and mesh
Based on the situations during the laser welding and mainly focused on the temperature
distribution, it is supposed that the laser resource is considered as a Gaussian distribution.
Also, on the basis of specimen size wire-cut, the calculating size is set as 25 mm (x) 20 mm
(y) 3 mm (z), the schematic of its finite element (FE) mesh is shown in Fig. 17. Moreover, Ti
filler is considered as a section of the substrate with the different properties to ignore the
effect of gap between the Ti filler and the substrate.

Welding Processes 68

Figure 17. FE mesh for 3D numerical analysis
4.2.3. Temperature distribution
The simulated results are shown in Fig. 18 to Fig. 23. It shows that the temperature without Ti
filler is same as the traditional laser welding. Simultaneously, due to the heat input into the
substrates directly, without the additional heat resource for melting Ti filler, the peak of
temperature (heat input) is relatively higher to form the weld. As a result, increasing the heat
input into the substrate will decrease the tensile strength of the welded joint and wide the heat
affected zone (HAZ) resulted in lower properties in the succedent practical applications (Table
2 and Fig. 19). Furthermore, a large amount of coarser acicular Al4C3 distributed in the fracture
surface as shown in Fig. 19, which decreased the tensile strength of the welded joints seriously.

Figure 18. Temperature distribution without Ti filler

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 69

Figure 19. Fractograph of the laser welded joint without Ti filler

Figure 20. Temperature distribution with Ti filler
Figure 20 shows the temperature field of laser welding SiCp/A356 with Ti filler. Considering
the Ti melting and in situ reaction in the welding pool as an endothermic reaction, the
welding temperature decreases and will be lower than that of laser welding directly (cf.
Figs. 18 and 20), and its temperature field is distributed more smoothly with in situ reaction
than that of laser welding without Ti filler as shown in Fig. 21. Also, the width of HAZ is
decreased to some extent (Fig. 21b). Furthermore, it shows that according to the real effect of
laser beam diameter, the thickness of Ti filler is about 0.3 mm will be optimal for in situ
welding which conformed to the experimental results as shown in Table 2.

Welding Processes 70

Figure 21. Temperature distribution of central heating on XOZ plane
(a) Temperature distribution on XOZ plane (b) Magnification of (a)
In addition, the effect of Ti on the temperature distribution on the central line is shown in
Fig. 22. It illustrates that the peak of the temperature is changed distinctly. Because of the
sandwiched Ti between the substrates and in situ endothermic reaction, the temperature of
substrate ahead of laser resource is lower than that of without Ti filler. Moreover, the
temperature at the succedent distance is increased or accumulated a little bit due to the
different conductive coefficient between Ti and substrate. On the other side, its
corresponding trend of the temperature behind the laser resource (resolidification) is same
as that of without Ti filler except for a peak appearance induced by more serious exothermic
potential during the crystallization.


Figure 22. Temperature distribution of central line on YOZ plane
(a) Temperature distribution on YOZ plane (b) Magnification of (a)
Figure 23 shows the temperature distribution when Ti filler is thick. The peak of
temperature is decreased obviously and leads to the welding failure.
(a)
(b)
(a) (b)

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 71

Figure 23. Temperature distribution with thick Ti filler

Figure 24. Microstructure and EDX of laser weld with thick Ti filler
(a)
(b
)

Welding Processes 72
Figure 24 shows the microstructure of laser welded joint with thick Ti filler and its
corresponding energy dispersive X-ray spectroscopy (EDX) results. It can be observed
that a large number of columnar Ti crystallization is distributed in the weld. From Figs.
23 and 24, it elucidates that with the increase of Ti thickness, the heat input into the
substrate is decreased and most of energy is used for melting Ti led to the insufficient in
situ reaction and stirring in the welding pool resulted in lower properties of welded
joints.
Furthermore, in order to verify the temperature field, noncontact thermometer (model
AZ9881) was used to measure the spot temperature on-line. The measured temperature
results are shown in Fig. 25. It shows that the measured results agree well with the
simulated results.




Figure 25. Surface temperature distribution in the processing center
4.3. Microstructure evolution simulation
According to the temperature calculation, the simulation of the evolution of the
microstructure based on thermodynamic equilibria, diffusion [5, 6, 7] was shown in Fig.
26. It showed that during the welding pool solidification, the in situ reinforcement
particles TiC would be formed around the initial reinforcement SiC particles. With the
increase of cooling time, the initial reinforcement SiC particles would be replaced by the
newly-formed in situ reinforcement particles TiC. It was well matched with the results
shown in Figs. 7 and 9.
Simulated results
Measured results

In situ Reaction During Pulsed Nd:YAG Laser Welding SiCp/A356 with Ti as Filler Metal 73

Figure 26. Simulation on microstructure evolution
(a) Initial (b) Start of solidification (c) Middle of solidification (d) End of solidification
5. Conclusions
The use of titanium as a filler metal in Nd:YAG laser welding of SiCp/A356 provided
beneficial in situ reinforcement effect. The effect of in situ reinforcement of the Ti filler
allowed the newly-formed reinforcement TiC particles to distribute uniformly in the weld
that subsequently resulted in successfully welding the SiCp/A356 composite. Moreover,
the typical pernicious interfacial reaction microstructure such as Al4C3 was effectively
restrained from the interface between aluminum matrix and reinforcement particles in the
Nd:YAG laser welding of SiCp/A356 with Ti filler. Furthermore, according to the
temperature calculation, the evolution of the microstructure was simulated based on
thermodynamic equilibria, diffusion. Results were well matched with the corresponding
experiments.
(a) (b)
(d)
SiC
TiC
SiC
SiC
TiC
(c)
TiC

Welding Processes 74
Author details
Kelvii Wei Guo
MBE, City University of Hong Kong, Hong Kong, China
Acknowledgement
The work is supported by a RGC general research fund (GRF) (Grant No.:9041503.) and a
Strategic Research Grant (SRG) from City University of Hong Kong (Grant No.: 7002287.)
6. References
[1] J. M. Gomez de Salazar, M. I. Barrena: Dissimilar Fusion Welding of AA7020/MMC
Reinforced with Al2O3 Particles. Microstructure and mechanical properties. Materials
Science and Engineering A 2003; 352(1-2): 162-168.
[2] W. Guo, M. Hua, H. W. Law and J. K. L. Ho: Liquid-Phase Impact Diffusion Welding of
SiCp/6061Al and Its Mechanism. Materials Science and Engineering: A, 2008, 490, (1-2),
427-437.
[3] W. Guo, M. Hua, and J. K. L. Ho: Study on Liquid-Phase-Impact Diffusion Welding
SiCp/ZL101. Compos. Sci. Technol., 2007, 67, (6), 1041-1046.
[4] L. M. Marzoli, A. V. Strombeck, J. F. Dos Santos, C. Gambaro, L. M. Volpone: Friction
Stir Welding of an AA6061/Al2O3/20p Reinforced Alloy. Composites Science and
Technology 2006; 66(2): 363-371.
[5] D. A. Porter, K. E. Easterling: Phase Transformations in Metals and Alloys, 2nd.
Cheltenham: Nelson Thornes, 2001.
[6] R. Riedel: Handbook of Ceramic Hard Materials. New York: Wiley-VCH, Weinheim,
2000.
[7] R. Boyer, G. Welsch, E. W. Collings: Materials Properties Handbook: Titanium Alloys.
Materials Park, Ohio: ASM International, 1994.
[8] J. R. Davis: ASM Specialty Handbook Aluminum and Aluminum Alloys. Materials
Park, Ohio: ASM International, 1993. p.557.
[9] S. Mall, T. Nicholas: Titanium Matrix Composites Mechanical Behavior. Lancaster, Pa.:
Technomic Pub. Co. Inc., 1998.
[10] W. M. Steen: Laser Material Processing, 3rd ed. London: Springer-Verlag; 2003.
[11] K. W. Guo: Influence of In Situ Reaction on the Microstructure of SiCp/AlSi7Mg
Welded by Nd:YAG Laser with Ti Filler. J. Materials Engineering and Performance,
2010, 19, 52-58.
[12] A.F. Mills: Heat and Mass Transfer. P. R. Donnelly & Sons Company; 1995.
Chapter 4




2012 Nawi et al., licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Nd:YAG Laser Welding
for Photonics Devices Packaging
Ikhwan Naim Md Nawi, Jalil Ali, Mohamed Fadhali and Preecha P. Yupapin
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/50656
1. Introduction
The state-of-the art of the pulsed Nd:YAG laser spot welding for photonics device
packaging has been introduced by Marley (Marley, 2002), which utilizes the laser for high
precision joining and alignment. The advantages of laser welding over conventional fusion
welding processes include precise welds with a high aspect ratio, narrow heat affected zone
(HAZ), very little thermal distortion, ease of automation, high welding speed, enhanced
design flexibility, clean, high energy density, low heat input and an efficient process (Zhou
& Tsai, 2008; Kazemi & Goldak, 2009). One of the key features of laser welding is the ability
to weld without filler materials and it offers distinct advantages (Pang et al., 2008). Laser
welding is a liquid-phase fusion process. It joins metals by melting the interfaces and
resulting the mixing of liquid molten metal. Then, it solidifies on the removal of the laser
beam irradiation (Ready, 1997). Photonics devices used for telecommunications of military
applications are usually required to operate for a long life of operation in fields with
potentially humid, corrosive and mechanically turbulent environments. Therefore, long
term reliability in such hostile operating conditions requires strong fixing of the aligned
components and hermetic sealing of the photonics devices inside metal hybrid housings
(Fadhali et al., 2007a). It is worth to mention that 60% to 80% of the photonics devices
modules cost are due to the coupling and packaging processes. Therefore, understanding
the effective packaging technique is very important to produce efficient and reliably
packaged photonics devices. Moreover, care must be taken to assemble functional packages
because only packages that can be manufactured reliably at competitive costs will survive in
the business world. For photonics packaging applications, most of the welds are of butt or
lap joints which require the weld penetration depth to be larger than the bead width.
Moreover, for miniature packages that contain some sensitive coupling components, the
penetration depth should be large enough to achieve a strong attachment (Fadhali et al.,

Welding Processes 76
2007a, 2007b, 2007c, 2007d, 2008). At the same time, the bead width should be small to
minimize the HAZ and hence prevent the damage to the sensitive optical components. The
desired material for this application requires a low thermal conductivity or a higher
electrical resistivity (Dawes, 1992). The lower the thermal conductivity of a material the
more likely it is to absorb laser energy. For this reason, several weldable grades of steel and
stainless steel are ideal for laser welding. The low carbon austenitic stainless steel (300 series
steel) which has carbon level less than 0.1% produces good quality welds and reliable weld
performance (Fadhali et al., 2007a; Naim et al., 2009). Hence, in this chapter, a stainless steel
304 is utilized as a base material for laser welding. It has been reported that other types of
stainless steel were studied by other researchers. For instance, Mousavi and Sufizadeh
investigated stainless steel 321 and 630 (Mousavi & Sufizadeh, 2008), while Beretta et al.
studied stainless steel 420 (Beretta et al., 2007), for the application of pulsed Nd:YAG laser
welding. Whilst a great deal of effort has been focused on developing processing systems
there is an urgent need to understand the strength of the weldment. It is the aim of this
chapter to examine the strength of a stainless steel 304 welded joint. Despite the pulsed
Nd:YAG laser welding has been widely used in microelectronic and photonics packaging
industry, a full understanding of various phenomena involved is still a matter of trials and
speculations.
2. Mathematical model of penetration depth of a spot weld
Penetration depth of laser spot welding is one of the vital parameters that contributes to a
good laser spot micro welding outcome. Practical penetration depth measurements are time
consuming and laborious like cutting the samples in order to obtain the weld cross section.
Thus, this section shows a theoretical model to estimate penetration depth inside the welded
samples by controlling laser parameters or to predict the laser parameters at the required
penetration depth. Figure 1 illustrates the penetration depth on a welding specimen. The 1-
Dimensional model is developed base on the heat conduction and energy balance equations
(Naim et al., 2010).

Figure 1. Illustration of the penetration depth of the focused laser beam on a welding specimen.

Nd:YAG Laser Welding for Photonics Devices Packaging 77
Applying the energy balance equation at the laser spot gives (Naim et al., 2010),

0
0
(1 )
m
z
l T
R I L K
t z
=
| | c c
=
|
c c
\ .
(1)
Here I0 is laser power density, Lm is latent heat of fusion, R is material reflectivity, is
material density, K is thermal conductivity, / T z c c is temperature gradient at the welding
front, l is the penetration depth, t is irradiation time and / l t c c is penetration velocity. To
determine the penetration velocity which is the change of penetration depth with time, the
temperature gradient at the welding front should first be determined. The temperature
distribution inside the target material is governed by the 1-Dimension heat conduction
equation given as (Naim et al., 2010),

2
2
1 T T
t
z
c c
=
o c
c
(2)
where, is thermal diffusivity. Equation (2) can be rewritten as

2
2
1 p
l
dT d T
t dz
dz
c
=
o c
(3)
The boundary conditions are taken as,
z=0, T=Tm and at z, T= T0 .
The maximum attainable temperature is the melting temperature of the material, Tm.
Substituting these boundary conditions into Equation (3) and the solution for temperature
distribution inside the material as

0
0
1
exp
m
T T dl
z
T T dt
(
| |
=
( |
o
( \ .

(4)
The temperature gradient at the welding front can be obtained using Equation (4),

( )
0
0
1
m
z
dl dT
T T
dz dt
=
| | | |
=
| |
o
\ . \ .
(5)
Substituting Equation (5) into the energy balance equation,

( ) ( )
0 0
1
m m
dl dl
R I L c T T
dt dt
| |
= +
|
\ .
(6)
From Equation (6), the welding velocity can be rewritten as,

( )
( )
0
0
1
m m
R I
dl
dt
L c T T

=
(
+

(7)

Welding Processes 78
Integrating Equation (7) gives,

( )
( )
2
0
1
o m m
R Pt
l
r L c T T

=
(
t +

(8)
Here P is the laser peak power, t is the irradiation time or pulse duration and r is the laser
beam radius. From Equation (8), one can observe that penetration depth is proportional to
laser peak power and pulse duration as well as the material reflectivity.
Figure 2 illustrates the penetration depth versus peak power and pulse duration. Any
changes of either peak power or pulse duration gives the same influence to the
penetration depth. This is in agreement with the fact that the penetration depth increases
linearly with both the peak powers and pulse durations. The penetration depth increases
linearly with an increase of laser peak power. The higher laser peak power means more
laser energy is absorbed into the welding material. When a higher laser energy is
absorbed, it will produce a deeper melting pool. After solidification, the melting pool
produces deeper weld or penetration depth. The trend is similar for variation of pulse
duration. Higher pulse duration means longer heating duration. This will provide more
time for laser beam to penetrate into the welding material. Thus, a deeper penetration
depth will be produced. From Figure 3, it can be summarized that the penetration depth
decreases significantly with laser spot radius. This is because the laser power density, I is
proportional to 1/r
2
and the power density will drop significantly as the laser spot radius
increases. Laser power density determines the quantity of energy applied on welding
material at a specific time. Higher power density will produce a deeper penetration
depth.

Figure 2. Penetration depth versus peak power and pulse duration.

Nd:YAG Laser Welding for Photonics Devices Packaging 79

Figure 3. Penetration depth versus peak power and laser beam radius.
3. Mathematical model of laser beam power penetration
In order to produce a spot weld, a laser beam is incident on a specimen surface. The incident
laser beam will be absorbed by the specimen depending on the material absorption
coefficient. Thus, the absorbed laser beam heats up the specimen surface, raising its
temperature. It will melt the specimen and produce a weld when it solidifies. The laser
beam penetration in welding material has been derived using the continuity equation (Naim
et al., 2010). It is assumed that laser beam penetrates the material based on a cylindrical
coordinate. Figure 4 illustrates the laser beam power penetration in stainless steel specimen.

Figure 4. Laser beam power penetration in stainless steel specimen.
The laser beam power penetration is derived from a continuity equation given as (Naim,
2010),

Welding Processes 80

0
P
k P
t
c
V =
c
(9)
In cylindrical coordinate, Equation (9) then becomes,

2 2
2 2
1 P P P P
k
t r r
r z
| |
c c c c
= + + |
|
c c
c c
\ .
(10)
Here P=P(r,z,t) is the laser beam power penetration, r is the laser spot radius, z is the depth, k
is the thermal conductivity and t is time. To solve this equation, let P = RZT = R(r)Z(z)T(t)
and the solutions are given by,

2
1
,
kt
T c e

= (11)

2 0 3 0
( ) ( ), R c J r c Y r = +

(12)
and
4 5
z z
Z c e c e

= + (13)
where c1, c2, c3, c4, c5 and are the unknown constants and determined by the boundary
conditions, J0 is first order of Bessel function and Y0 is second order of Bessel function.
Equation (10) can be solved by using Equations (11), (12) and (13) giving,

2
1 2 0 3 0 4 5
( , , ) [ ][ ( ) ( )][ ]
kt z z
P r z t c e c J r c Y r c e c e

= + + (14)
Let A= c1= c2= c3= c4= c5 as the incident laser beam peak power and is the material
absorption coefficient. In this calculation, only first order of Bessel function is considered. It
is also assumed that the laser beam propagates only directing into the material which is z
direction. Equation (14) can be written as,

2
0
( , , ) [ ( )][ ]
k t z
P r z t Ae J r e

= (15)
For stainless steel material, the absorption coefficient, is 0.3 (Kazemi & Goldak, 2009) and
the laser beam peak power used in this consideration is 3.5kW. Equation (15) then specified as,

2
0.3 0.3
0
( , , ) 3500 [ (0.3 )][ ]
kt z
P r z t e J r e

= (16)
Equation (16) provides the time dependent laser beam penetration. The laser beam penetration
is computed for its radius and depth. Figure 5 shows the profile of applied peak power versus
the depth of specimen and laser spot radius. The peak power 3.500kW decreases exponentially
with depth of stainless steel material and decreases according to the first order of Bessel
function in terms of the spot radius. The peak power incident on the surface is 3.500kW at the
centre of laser spot target, but decreases to 3.0248 kW at the spot radius of 0.25mm. Peak
power reduces to 0.1743kW at the centre and 0.1506kW at spot radius of 0.25mm for a depth of
1.00mm. At the laser spot radius of 2.00mm, the laser beam is almost fully penetrated with
only peak power of 0.0087kW at the centre of the penetration.

Nd:YAG Laser Welding for Photonics Devices Packaging 81
Figure 6 illustrates the time dependent laser beam penetration for stainless steel material.
According to the illustration, the peak power reduces exponentially for both depth and time.
From observation, laser beam peak power decreases faster through the depth rather than
through the time, relatively. This is explained by Equation (16) which states penetrated laser
beam is based on absorption coefficient through the depth is -0.03 m but only -0.0009 m
through time. At a depth of 2.00 mm and time of 4.0 ms, the peak power is 0.0002 kW. At
this point, laser beam is almost fully penetrated. On the interface between stainless steel
material and laser beam spot, laser beam penetration can be illustrated by Figure 7 at certain
time. As discussed before, laser beam penetration decreases according to the first order of
Bessel function with the spot radius and exponentially with time. At the central of the
interface, the peak power is 3.5 kW and it reduces to 0.0026 kW after 8.0 ms. At laser spot
radius of 0.25 mm, the laser beam penetration is 3.0248 kW and it reduces to 0.0023 kW after
8.0 ms where the laser beam is about to fully absorbed by the stainless steel material.

Figure 5. Peak power versus depth and radius for stainless steel with absorption coefficient, =0.3 and
t=1 ms.

Figure 6. Peak power versus depth and time for stainless steel with absorption coefficient, =0.3 at the
centre of penetration.

Welding Processes 82

Figure 7. Peak power versus radius and time for stainless steel with absorption coefficient, =0.3 at the
welding front.
4. Analysis of the pulsed Nd:YAG laser spot weld
The profiles for the penetration depth and bead width produced by the pulsed Nd:YAG
laser beam are depicted in Figure 8 and Figure 9. The weld dimensions (weld penetration
depth and bead width) for different laser beam peak powers are illustrated in Figure 10. The
results display an increase in penetration depth and bead width with an increase in the laser
beam peak power. The deepest penetration depth produced is 1.31 mm and the largest bead
width is 0.57 mm when laser beam peak power is set at 3.5 kW. When laser beam peak
power is reduced to 0.5 kW, the reading of penetration depth and the bead width are only
0.36 mm and 0.24 mm. The linear gradients of penetration depth and bead width are 0.301
and 0.10, respectively. These values show the laser beam peak power is almost 3 times more
effective on the penetration depth rather than the bead width. This suggests that the laser
beam peak power is a reliable parameter to control the desired penetration depth. It is
observed in Figure 10 that the penetration depth and bead width increases when the pulse
duration is increased. From the Figure 11, linear gradient of penetration depth and bead
width are 0.0039 and 0.035, respectively. Only slight difference is noted and this means that
the pulse duration has no significant effect either on penetration depth or bead width. As
compared with the effect of laser peak power, the pulse duration is a much more better
parameter to control the desired bead width.
Laser beam can be defocused by moving the focus point position forward or backward from
the specimen surface. Figure 12 depicts that the penetration depth decreases significanly
when the laser beam focus point position is moved from 0 to 4.0 mm with respect to the
specimen surface. After 4.0 mm, the penetration depth decreases gradually untill 6.0 mm
and there is no penetration depth that can be traced after 6.0 mm. Otherwise, when the focus
point position is moved forward or backward, the bead width increases. But after 6.00 mm,
there is no more bead width and only the sign of burning can be observed on the welding

Nd:YAG Laser Welding for Photonics Devices Packaging 83
surface. As the focus point position is moved forward or backward from the specimen
surface, the laser spot size increases. This will reduce the intensity of the laser beam which is
given by, I0=P/ r
2
. Here, P is the laser beam peak power and r is the laser spot radius. It
indicates that the intensity of laser beam is negatively proportional with r
2
. Hence, the laser
beam does not have sufficient intensity to penetrate the material. As illustrated in Figure 13,
it is observed that the penetration depth and bead width changes with the laser beam
incident angle. As the laser beam incident angle increases, the laser spot becomes elliptical
and wider. Hence the bead width also becomes elliptical and wider. The wider laser spot
will reduce the intensity of the laser beam. This will result in a decrease of the penetration
depth. However, when the laser beam incident angle is increased, the reflectivity of material
surface drops due to the influence of light polarization. When the reflectivity drops, it will
transport more laser energy into the welding material. Hence, it will increase the
penetration depth. It can be seen that when weld penetration depth at angle of incidence of
45 degrees is much higher than that at 25 degrees. But, even though the reflectivity of the
material surface drops greatly at 65 degrees, the penetration depth is only 0.82 mm because
laser beam is less intense due to the significant increases in the laser spot area for a slight
increase of the incident angle. In Figure 14, it can be seen that the penetration depth
increases slightly with the number of shots. The penetration depth for the first and seventh
pulse shots are 0.91 mm and 1.11 mm. When the first shot is applied it produces a shallow
concave hole on the specimen surface. This is due to the material ablation produced by laser
pulse pressure when it strikes the material surface. Then, when the second shot is given, it is
able to go towards the bottom of the concave hole. The second shot weld penetration depth
is similar to the first shot but with an increase in the depth of the concave hole. The first and
seventh shot produce 0.52 mm and 0.57 mm of bead width. The difference is relatively small
because the same laser beam parameters are used for each shot.

Figure 8. Cross-section of a spot weld produced by 2.5kW laser beam peak power and 2.5ms pulse
duration. The penetration depth is labelled by the vertical line.

Welding Processes 84

Figure 9. Figure 9. Top view of a spot weld produced by 2.5kW laser peak power and 2.5ms pulse
duration. The bead width is labelled by the horizontal line.

Figure 10. Characteristics of a spot weld dimensions as a function of laser peak powers conducted by
2.5 ms pulse duration with laser beam incidence is vertical in respect to the surface normal.

Figure 11. Characteristics of a spot weld dimensions as a function of laser pulse durations conducted
by 2.5 kW laser peak power with wavelength of 1064 nm.

Nd:YAG Laser Welding for Photonics Devices Packaging 85

Figure 12. Characterictics of a spot weld dimensions when Nd:YAG laser spot is positioned forward and
backward from the stainless steel 304 surface with 2.5 kW laser peak power and 2.5 ms pulse duration.

Figure 13. Characterictics of a spot weld dimensions when Nd:YAG laser spot incident angle is varied
with respect to the stainless steel 304 surface normal by employing 2.5 kW laser peak power and 2.5 ms
pulse duration.

Figure 14. Spot weld produced by 2.5 kW laser peak power, 2.5 ms pulse duration with laser beam
incidence is vertical in respect to the surface normal and laser focus point is on the stainless steel 304
surface, while the laser beam shot numbers are varied.

Welding Processes 86
5. The strength analysis of a welding joint
In this section, the strength test of a welding joint of a single spot welding is discussed for
stainless steel 304. A test is also executed for a seam welding which is produced by shooting
the laser pulses continuously along the 10.0 0.1 mm stainless steel 304 interfaces. The
strength for seam welding of stainless steel 304 is also compared to an Invar
TM
, which is a
commercial welding material for photonics device packaging. A Unitek Miyachi LW10E
ultra compact pulsed Nd:YAG laser with wavelength of 1.064 m is employed to produce a
weld joint. The energy per pulse output is in the range of 1 to 20 J, laser beam peak power is
up to 3.5 kW and pulse duration ranging from 0.3 ms to 10.0 ms. However, in this
investigation, only 3.5 kW laser beam peak power and 6.5 ms pulse duration is employed.
This parameter is chosen because it produces a deepest penetration depth on stainless steel
304 which is 1.2 mm (Nawi et al., 2011). The penetration depth is important and should be
large enough to achieve a strong attachment. The welding base material used is a stainless
steel 304 sheet with a thickness of 1.00.1mm. The experimental setup to produce a weld
consists of a laser source, fiber optics delivery system, and focusing lenses with focal length
of 100.00 mm as illustrated in Figure 14.
The optical fiber cables transmit the laser beam to the welding focusing lenses inside the
lens housing. The laser welding system is also equipped with an aiming diode laser beam,
which simplifies positioning of the laser spot on the stainless steel sheet. A spot weld and a
seam weld are produced with two types of joint; butt joint and lap joint, as illustrated in
Figures 15(a) and 15(b). The strength of the weld joints is conducted by pulling the joint
materials using INSTRON Series IX/s Automated Materials Tester System. The crosshead
speed utilized for pulling the joint is set to 0.2 mm/min. The strength for a weld joint is
determined by measuring the applied load during the test.

Figure 15. Nd:YAG laser experimental setup to produce a spot weld on stainless steel 304 specimen.

Nd:YAG Laser Welding for Photonics Devices Packaging 87
The element composition of stainless steel 304 and welded stainless steel 304 are examined
by using Energy Dispersive X-ray (EDX) technique. The stainless steel 304 consists of iron
(Fe) 69.59wt%, chromium (Cr) 18.33wt% and nickel (Ni) 12.08wt%. Figure 16 shows the
element composition for stainless steel 304 changes when it is welded by Nd:YAG laser
beam and it is now comprising of iron (Fe) 68.83wt%, chromium (Cr) 17.10wt% and nickel
(Ni) 14.08wt%. It is observed that there are no changes in the elemental composition before
and after welding of stainless steel 304. The applied load for the test and the pulling
displacement for spot and seam welding are shown in Figure 17 and Figure 18, respectively.
Maximum applied load is considered as the strength of the weld joint. Figure 16 illustrates
maximum load for butt and lap joint which are 0.2296 kN and 0.0691 kN with a pulling
displacement of 0.43 mm and 0.40 mm, respectively.

Figure 16. Configuration of the (a) butt joint and (b) lap joint, for stainless steel 304 with a thickness
1.00.1mm.

Figure 17. EDX analysis for the elemental composition of welded stainless steel 304.
The Invar
TM
maximum load is recorded as 0.3760 kN with 0.43 mm pulling displacement.
The outcome indicates that the butt joint produces much stronger attachment than the lap
joint. The stronger attachment is contributed by a larger volume of welded stainless steel
at the joint interfaces as shown in Figure 19(a). For lap joint, the thickness of the stainless
steel 304 becomes double as the upper and lower stainless steel sheets are overlapped. If

Welding Processes 88
the laser beam produces insufficient penetration depth, thus only a small volume of
welded stainless steel 304 is produced at the interfaces of the joint as shown in Figure
19(b). This will lead to a weaker joint. But, the penetration depth can be increased by
controlling the laser beam parameters such as laser beam peak power and pulse duration.
Hence, it will increase the strength of a lap joint. In terms of welding materials
comparison, the commercial welding material, Invar
TM
, provides more robust joint than
stainless steel 304. Figure 18 indicates the seam welding maximum load for butt joint is
2.6339 kN with 0.62 mm pulling displacement. Lap joint recorded the maximum load of
1.0466 kN at 0.49 mm. This implies that the butt joint of a seam welding provides a much
stronger attachment than the lap joint similar with the case of a single spot weld. Figure
18 also indicates between the butt joint pulling displacement of 0.48 mm and 0.72 mm,
there are no significant changes in the joint strength because each point along the seam
welding has the similar maximum strengths but it is not achieved simultaneously
depending on its geometry.

Figure 18. Strength profile of a single spot weld produced on stainless steel 304 and Invar
TM
by using
3.5 kW Nd:YAG laser beam peak power and 6.5 ms pulse duration.

Figure 19. Strength profile of a seam weld produced on a stainless steel 304 along 10.0 0.1 mm joint
interfaces.

Nd:YAG Laser Welding for Photonics Devices Packaging 89

Figure 20. FESEM micrographs of a stainless steel 304 spot weld joint; (a) butt joint and (b) lap joint,
produced by 3.5 kW laser beam peak power and 6.5 ms laser pulse duration.

Figure 21. The strength comparison between butt joint and lap joint for variation number of spot weld.

Figure 22. The strength of a stainless steel 304 with 1.0 0.1 mm thickness and 10.0 0.1 mm width.
a) b)

Welding Processes 90
In miniature assemblies, only a small number of laser spot weld is used for an attachment
because of the relative small parts (Fadhali et al., 2007a, 2007b, 2007c, 2007d, 2008). The
strength on the number of laser spot welding up to four spots is investigated for butt joint
and lap joint. Then, the maximum strength for both joint types is compared. Figure 20
illustrates that the strength increases when the number of spots increases for butt and lap
joint. A linear relationship is obtained from both joints strength against the number of spots
and it has equal gradient which are 0.122 and 0.120, respectively. This suggests there is no
significant difference in the influence of number of spots on the strength for both types of
joint. The strength of stainless steel 304 is shown in Figure 21. It produces a strength of
5.2866 kN with 18.23 mm pulling displacement. The strength of stainless steel 304 is almost
twice stronger than the seam welded stainless steel 304. Thus the welded material cannot
attain the strength of the base material itself. The pulling displacement indicates that the
elastic modulus of stainless steel 304 is much more greater than the welded stainless steel
304.
6. Conclusion
In conclusion, it is shown that the penetration depth and bead width increase when the laser
beam peak power, pulse duration and number of shot are increased. It is found that the laser
peak power is more effective to produce the deeper penetration depth rather than the bead
width. Otherwise, the laser beam pulse duration is an accurate parameter to control if the
desired bead width is required rather than the penetration depth. As the focus point
positions are placed forward and backward from the stainless steel 304 specimen surface,
the laser spot size increases. This leads to the reduction of the laser beam intensity. Hence,
the penetration depth decreases. The increase of the laser spot size will increase the bead
width. When the laser beam incident angle is varied, the bead width also changes due to the
widening of the laser spot size. The penetration depth depends on the both widening laser
spot size and material surface reflectivity that drops due to the influence of light
polarization. The penetration depth increases slightly with the number of shots due to the
material ablation produced by laser pulse pressure when it strikes the specimen surface. For
welding joint strength, a butt joint produces much stronger attachment than a lap joint. The
stronger attachment is contributed by a larger volume of welded stainless steel 304 at the
joint interfaces. It is observed that each point along the seam welding has a similar
maximum strengths but it is not achieved simultaneously. The similar gradient obtained in a
linear relationship suggests that there is no significant difference in the influence of number
of spots on the butt and lap joint strength. The commercial welding material, Invar
TM
,
provides more robust joint rather than stainless steel 304. The welded stainless steel 304
cannot attain the strength and elastic modulus of the stainless steel 304 itself. From the
strength test results, it is found that stainless steel 304 is a good candidate to be used as a
welding material for photonics device packaging by using pulsed Nd:YAG laser welding
technique.

Nd:YAG Laser Welding for Photonics Devices Packaging 91
Author details
Ikhwan Naim Md Nawi
Centre of Foundation Studies & Faculty of Applied Sciences, Universiti Teknologi MARA, Malaysia
Jalil Ali
Universiti Teknologi, Malaysia
Mohamed Fadhali
Physics Department, Faculty of Science, Ibb University, Yemen
Preecha P. Yupapin
King Mongkuts Institute of Technology Ladkrabang, Bangkok, Thailand
7. References
Beretta, J.R.; Rossi, W.; Neves, M.D.M.; Almeida, L.A. & Vieira, N.D. (2007). Pulsed Nd:YAG
laser welding of AISI 304 and 420 Stainless Steel. Journal of Optical Lasers Engineering,
Vol. 45, pp. 960-966.
C. Dawes, C. (1992). Laser welding: A practical guide. Woodhead Publication Ltd., England.
Fadhali, M.; Zainal, J.; Munajat, Y.; Ali, J. & Rahman, R. (2007a). Reliable pigtailing of
photonic devices employing laser microwelding. Journal Engineering and Applied Science,
Vol. 2, pp. 1724-1728.
Fadhali, M.; Zainal, J.; Munajat, Y.; Ali, J. & Rahman, R. (2007b). Investigation of the
application of Nd:YAG laser welding to couple photonic devices and packaging. Lasers
Engineering, Vol. 17, pp. 273-286.
Fadhali, M.; Zainal, J.; Munajat, Y.; Ali, J. & Rahman, R. (2007c). Laser diode pigtailing and
packaging using Nd:YAG laser welding technique. Jurnal Komunikasi Fisika Indonesia,
Vol. 5, pp. 203-208.
Fadhali, M.; Zainal, J.; Munajat, Y.; Ali, J. & Rahman, R. (2007d). Analysis of laser
microwelding applied for photonic devices packaging. Journal Solid State Science and
Technology Letter, Vol. 14, pp. 18-19.
Fadhali, M. (2008). Efficient coupling and reliable packaging of photonic devices using laser
welding technique. Ph.D Thesis: Universiti Teknologi Malaysia, Skudai, Johor, Malaysia.
Kazemi, K & Goldak, J.A. (2009). Numerical simulation of laser full penetration welding.
Computer Material Science, Vol. 44, pp. 841849.
Marley, C. (2002). Laser welding photonics devices: Proper design guidelines enable users to
achieve high-precision alignment requirements. Industrial Laser Solutions for
Manufacturing, 2002.
Mousavi, S.A.A. & Sufizadeh, A.R. (2008). Metallurgical investigations of pulsed Nd:YAG
laser welding of AISI 321 and AISI 630 stainless steels. Journal of Material Design, Vol. 15,
pp. 44-49.
Naim, I.; Saktioto, T.; Fadhali, M.; Ali, J.; & Yupapin, P.P. (2009). An investigation of a
pulsed Nd:YAG laser welding technique. Proceedings of IEEE-Regional Symposium on
Micro and Nanoelectronics, pp. 274-280, Kelantan, Malaysia, August 10-12, 2009.

Welding Processes 92
Naim, I.; Saktioto, T.; Hamdi, M.; Fadhali, M.; Ali, J.; & Yupapin, P.P. (2010). Penetration
depth estimation for stainless steel 304L using Nd:YAG laser spot micro welding.
Proceedings of 2
nd
Topical Meeting on Lasers and Optoelectronics, Vol. 1, pp. 278-288,
Terengganu, Malaysia, March 13-15, 2010.
Nawi, I.N.; Saktioto; Fadhali, M.; Hussain, M.S.; Ali, J. & Yupapin, P.P. (2011). A Reliable
Nd:YAG Laser Welded Stainless Steel 304 for Photonics Device Packaging. Procedia
Engineering, Vol. 8, pp. 380-385.
Pang, M.; Yu.G.; Wang, H.H. & Zheng, C.Y. (2008). Microstructure study of laser welding
cast nickel-based superalloy K418. Journal of Material Processing Technology, Vol. 207, pp.
271275.
Ready, J.F. (1997). Industrial applications of lasers. New York: Academic Press.
Zhou, J. & Tsai, H.L. (2008). Modeling transport phenomena in hybrid laser-MIG keyhole
welding. Internattonal Journal of Heat and Mass Transfer, Vol. 51, pp. 4353-4366.
Chapter 5




2012 El-Batahgy, licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Laser Beam Welding of Austenitic Stainless
Steels Similar Butt and Dissimilar Lap Joints
Abdel-Monem El-Batahgy
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/48756
1. Introduction
1.1. Laser beam welding of similar butt joints of austenitic stainless steels
Because of its inherent corrosion resistance, austenitic stainless steels, known as 300 series,
have become cost-effective, staple materials for long-term applications in many industrial
sectors including gas, petroleum, petrochemicals, fertilizers, food processing, and pulp
industries as well as power generating plants. They have found also widespread use for
manufacturing of chemical installations including stationary pressure tanks and tanks for
transport of liquid and compressed gases, pipelines of high diameter in water power plants,
for manufacturing of ships for transport of chemicals and installations of drilling rigs, etc.
Thick-section stainless steels are widely used in the components and structures for nuclear
power plants.
For all applications of austenitic stainless steels, welding is of considerable importance since
it is widely used in components' manufacturing. In comparison with ferritic steels, lower
thermal conductivity and higher thermal expansion coefficient of austenitic stainless steels
results in larger thermal distortions and internal stresses of the welded parts, which increase
susceptibility of the weld to hot cracks.
Another possible welding problem of austenitic stainless steels is sensitization that occurs at
900-1400F (482-760C) during cooling after welding where chromium carbides form along
the austenite grains and causes depletion of chromium from the grains resulting in
decreasing the corrosion protective passive film.
In this concern, austenitic stainless steels poses distinct challenges when it is joined with gas
tungsten arc welding (GTAW) due higher possibility of carbide precipitation and distortion
in comparison with laser welding. In other words, joining austenitic stainless steels with

Welding Processes 94
GTAW can be tricky, but with a laser, it can be done successfully. Previous studies of the
weldability of stainless steels indicate that the basic condition for ensuring high quality of
welded joints and reducing thermal distortions to minimum is reducing the heat input of
welding that is ensured only by laser welding.
CO2 laser beam welding with a continuous wave, which is widely used for stainless steels
components, is a high energy density and low heat input process. The result of this is a
small heat-affected zone (HAZ), which cools very rapidly with very little distortion, and a
high depth-to-width ratio for the fusion zone.
The heat flow and the fluid flow in the weld pool can significantly influence the temperature
gradients, the cooling rates and the solidification structure. In addition, the fluid flow and
the convective heat transfer in the weld pool are known to control the penetration and shape
of the fusion zone [1].
Generally, laser beam welding involves many variables; laser power, welding speed,
defocusing distance and type of shielding gas, any of which may have an important effect
on heat flow and fluid flow in the weld pool. This in turn will affect penetration depth,
shape and final solidification structure of the fusion zone. Both the shape and
microstructure of the fusion zone will considerably influence the properties of the
weldment.
There are many reports [2-4] that deal with the shape and solidification structure of the
fusion zone of laser beam welds in relation to different laser parameters. However, the effect
of all influencing factors of laser welding has up to now not been extensively researched.
More work is required for understanding the combined effect of laser parameters on the
shape and microstructure of the fusion zone.
The present investigation is concerned with laser power, welding speed, defocusing
distance and type of shielding gas and their effects on the fusion zone shape and final
solidification structure of some austenitic stainless steels.
1.2. Experimental procedure
Three types of commercial austenitic stainless steels, 304L, 316L and 347, were used. Their
chemical composition and mechanical properties are given in Table 1. The thickness of both
304L and 316L steels was 3 mm while that of 347 steel was 5 mm.


Table 1. Chemical composition (wt%) and mechanical properties of the used base metals

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 95
Both bead-on-plate and autogenous butt weld joints were made using a carbon dioxide laser
capable of producing a maximum output of 5 kW in the continuous wave mode. Bead-on
plate was made on plates with 3 mm thickness while autogenous butt weld joints were
made on plates with 3 and 5 mm thickness. Specimens with machined surfaces were
prepared as square butt joints with dimensions of 125x150 mm and were held firmly using
fixture to prevent distortion
The laser beam welding parameters investigated are summarized in Table 2. Combinations
of laser power (P) of 2-5 kW and speed (S) of 0.5-3 m/min resulted in nominal heat in
nominal heat inputs (HI) ranging from 0.04 to 0.48 kJ/mm. The defocusing distance (Dd) was
in the range of -5 to 3 mm. Shielding was made using either argon or helium gas.


Table 2. Welding parameters used
After welding, the specimens were visually inspected then, sectioned transverse to the
welding direction. The shape and microstructure of the fusion zone were examined using
optical microscopy. Micro-compositional analysis of welds was performed using an electron
probe micro-analyzer (EPMA) at an accelerating voltage of 25 kV.
Mechanical tests including tensile, bending and hardness measurements of butt welds
having complete penetration were performed according to relevant standards. The data
reported are the average of three individual results.
1.3. Results and discussion
1.3.1. Macrostructure of laser beam welds
1.3.1.1. Effect of laser power
The effect of heat input as a function of laser power, HI = P/S, was clarified using type 304L
and type 316L steels. Both welding speed and defocusing distance were kept constant at 3
m/min and zero respectively.

Welding Processes 96
The penetration depth increased sharply with increasing laser power from 2 to 3 kW as
shown in Figure 1. Complete penetration for the 3mm base metal was obtained at laser
power equal to or greater than 4 kW. Figure 9 shows an example of a cross section of type
304L steel butt weld made using laser power of 4 kW. The weld bead showed a
characteristic of laser welding with dept / width ratio close to 3. No welding cracks or
porosity were found in any of the welds, this may be partly due to the good crack resistance
of the base metal and the welding conditions provided.
The results indicated also that the development of the weld pool is essentially symmetrical
about the axis of the laser beam. Yet, lack of symmetry at the root side was observed
particularly at higher welding speed (Figure 2) suggesting an unsteady fluid flow in the
weld pool. This is due to the presence of two strong and opposing forces, namely, the
electromagnetic and the surface tension gradient forces. At these locations, the
electromagnetic force may have overcome the surface tension force, thereby, influencing
convective heat transfer. As a result, any local perturbation in the weld pool can cause the
flow field to change dramatically, resulting in the observed lack of local symmetry.

Figure 1. Effect of laser power on penetration depth of types 304L and 316L stainless steel welds.

Figure 2. A cross section of type 304L stainless steel weld made using P=4kW, S=3m/min, Dd=0.0mm.

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 97
Laser power has a less influence on both weld profile and HAZ width in comparison with
its effect on penetration depth. This is in agreement with other researchers work where they
pointed out that changing laser power between 3 and 5 kW [5] did not result in any
significant change in the size or shape of the weld.
It is expected that similar results concerning the dependence of penetration depth on laser
power could be obtained in the case of type 347 steel due to similarity in both physical and
mechanical properties. The optimum power for complete penetration with acceptable weld
profile for the 5mm base metal thickness was 5 kW at a welding speed of 2 m/min as shown
in Figure 3.

Figure 3. Optimum weld profile of type 347 stainless steel made using P=5kW, S=2m/min, Dd=-0.4mm.
1.3.1.2. Effect of welding speed
The effect of welding speed was investigated at the optimum laser power (4 kW) and zero
defocusing distance. Figure 4 shows the relationship between welding speed and fusion
zone depth/width ratio for both 304L and 316L base metals. The depth/width ratio increased
sharply from 2.1 to 4.1 with the increase in welding speed from 0.5 to 3 m/min.
The dependence of depth/width ratio on welding speed was confirmed at a different laser
power (3 kW). A lower welding speed resulted in a considerable increase in the fusion zone
size and consequently a decrease in depth/width ratio leading to unacceptable weld profile.
Complete penetration with relatively acceptable fusion zone size for the 3mm base metal
thickness was obtained at welding speed of 2 m/min as shown in Figure 5. The fusion zone
is symmetrical about the axis of the laser beam.
The above results have shown that the laser power and welding speed should be optimized
in order to minimize heat input, then a satisfactory weld with reliable quality could be
obtained. This reflects one of the most notable features of laser welding compared with
other welding processes, which is small heat input.
Turning to the macrographs shown in Figures 5 and 8, complete penetration with relatively
acceptable fusion zone profile could be obtained using either 4 kW, 3 m/min (Figure 5) or 3

Welding Processes 98
kW, 2 m/min (Figure 8). However, 4 kW, 3 m/min resulted in a smaller fusion zone size with
less inflection at its interface in addition to the high welding speed in this case.


Figure 4. Influence of welding speed on weld depth/width ratio of types 304L and 316L stainless steels.

Figure 5. A cross section of type 304L stainless steel weld made using P=3kW, S=2m/min, Dd=0.0mm.
At high welding speed, attenuation of beam energy by plasma is less significant. This results
in relatively more exposure of the laser beam on the sample surface. Consequently, the
depth/width ratio would be increased and the fusion zone size would be minimized.
1.3.1.3. Effect of defocusing distance
Defocusing distance, focus position, is the distance between specimen surface and the
optical focal point. In order to study its effect on both penetration depth and weld profile,
bead-on-plate was made with changing defocusing distance between -5 and 3 mm. Low
laser power (2 kW) and high welding speed (3 m/min) were selected to obtain incomplete
penetration.

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 99
Examples of weld cross sections of type 304L steel made using different defocusing
distances are shown in Figure 6. No cracking or porosity was observed in all welds. The
penetration depth is considerably decreased with changing defocusing distance from zero
(Figure 6-b) to either minus (Figure 6-a) or plus (Figure 6-c) values as a result of decreasing
laser beam density.
The relationship between defocusing distance and penetration depth of both 304L and 316L
steels is summarized in Figure 7. The penetration depth decreased from 1.9 to 1.6 mm on
changing the defocusing distance from zero to either -1 or 1mm. Then, the penetration depth
decreased sharply to about 0.2 mm on changing the defocusing distance to more negative (-5
mm) or positive (4 mm) values.
These results indicated that the most effective range of defocusing distance to get maximum
penetration with acceptable weld profile lies between zero and -1 mm. In order to obtain the
optimum value, complete penetration butt welds were made using previously obtained
optimum laser power (4 kW) and optimum welding speed (3 m/min). The most acceptable
weld profile was obtained at defocusing distance of - 0.2 mm for 3 mm thickness where
weld bead depth/width ratio is maximum and fusion zone size is minimum with a slight
taper configuration as shown in Figure 8. However, the optimum defocusing distance to
attain acceptable weld profile for 5 mm thickness was -0.4 mm (Figure 3). The smooth
curved and symmetrical fusion zone interface shown in Figures 3 and 8 suggests that the
driving forces for fluid flow in the weld pool, buoyancy and surface tension gradient,
augment each other, resulting in a coherent flow field.

Figure 6. Cross section of type 304L stainless steel welds made using P=2kW, S=3m/min with different
defocusing distances. (a) Dd=-3.0mm, (b) Dd=0.0mm, (c) Dd=+2.0mm.

Welding Processes 100


Figure 7. Relationship between defocusing distance and penetration depth of types 304L and 316L
stainless steels.


Figure 8. A cross section of type 304L stainless steel weld made using P=4kW, S=3m/min, Dd=-0.2mm.
1.3.1.4. Effect of type of shielding gas
In all previous experiments, argon was used as a shielding gas. For comparison, argon was
replaced by helium while other laser parameters were kept constant. Weld profile is
remarkably improved where fusion zone interfaces are almost parallel to each other as
shown in Figure 9.
In general, when the laser beam interacts with the workpiece, a hole is drilled through the
thickness of the material. This hole or cavity is filled with a plasma and surrounded by
molten metal, thus, the high energy density of the focused beam could be lost easily. This
plasma effect was reduced as a result of the higher ionization potential of helium then, the
weld profile was improved.

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 101

Figure 9. Optimum weld profile of type 304L stainless steel obtained using helium as a shielding gas
instead of argon with the same laser parameters as of Figure 15.
1.3.2. Microstructure of laser beam welds
Microstructures of type 304L steel weld metals made using two different welding speeds, 1
and 3 m/min, with same laser power, 4 kW, are shown in Figure 10. The noticeable feature is
the highly directional nature of the microstructure around the axis of the laser beam. This is
due to solidification of the weld metal at high cooling rate compared to that of conventional
GTA welding [6]. It can also be noticed that the higher the welding speed, the finer the
dendritic structure (Figure 10-b). This is attributed to an increase in both solidification and
cooling rates due to low heat input resulted from high welding speed.

Figure 10. Optical micrographs of weld metals of type 304L stainless steel made using P=4kW with
different welding speeds. (a) S=1m/min, (b) S=3m/min.
Concerning the effect of laser power, the higher the laser power, the coarser is the dendritic
structure due to decreasing cooling rate. However, the effect of laser power was relatively
less than that of welding speed.

Welding Processes 102
The microstructures of all laser beam welds were always austenite with a few percent of
delta-ferrite at the dendritic boundaries. The existence of delta-ferrite was confirmed by
both colour etching and electron probe microanalysis (EPMA). The amount of delta-ferrite
was estimated using Cr and Ni equivalents [7] of weld metal chemical composition. Based
on the Schaeffler diagram, about 2 or 3 vol% ferrite was expected to exist in the austenitic
matrix. It should be reported that no weld solidification cracking was observed in any of the
welds evaluated.
Under normal weld solidification conditions, the solidification mode in austenitic stainless
steels is primarily a function of composition, with a shift from primary ferrite to primary
austenite accomplished by reducing the Creq/Nieq ratio below 1.5 [8].
However, Suutala [2], Vitek and David [9], and Lippold [10] have illustrated that the
boundary between primary austenite and primary ferrite solidification is not just a function
of weld metal composition, but is a function of the growth rate.
The subject results are consistent with the modified Suutala diagram [11] for predicting
microstructure and cracking of austenitic stainless steels under rapid weld solidification
conditions encountered during laser welding. All the steels tested exhibited a Creq/Nieq ratio
greater than 1.7, the value suggested by the modified Suutala diagram as the demarcation of
crack susceptibility. Consequently, it can be deduced that all welds concerned in this
investigation were solidified as mixed mode of primary and massive austenite. The
transition in primary solidification from ferrite to austenite could be attributed to weld pool
undercooling as a result of extremely high solidification growth rate [12-15].
The results are also in agreement with Lippold [12] work where he pointed out that cracking
in austenitic stainless steel welds is avoided as the proportion of primary ferrite in the
mixed mode solidification increases.

Figure 11. Hardness profiles of base metal, HAZ and weld metal of type 304L stainless steel as a
function of heat input.

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 103
1.3.3. Mechanical properties
Tensile test results of all laser butt welds with complete penetration showed that failure has
taken place in the base metal. The bending test at room temperature showed no cracks in all
joints as a result of high ductility.
Typical hardness profiles of the base metal, HAZ and weld metal of type 304L steel as a
function of heat input are shown in Figure 11. No significant difference between hardness of
the base metal and that of weld metal or HAZ was obtained. Hardness of both weld metal
and HAZ was slightly higher than that of the base metal regardless of heat input. These
results were also valid for the other two base metals. This is expected because mechanical
properties of steel, in general, are based on its microstructure.
2. Laser beam welding of austenitic stainless steel in lap joints with zn-
coated carbon steel
2.1. Introduction
Because of its excellent corrosion resistance, austenitic stainless steel has found widespread
use in the paper making equipment includes pressure vessels, storage tanks, piping, hopper,
bins, chutes and structural components. For all of these applications, attachments such as
access platforms, catwalks, stiffeners, column supports, stairways, washers and pipe
hangers are welded to the outside surfaces of the equipment. Zn-coated carbon steel is often
specified for these attachments due to its good corrosion resistance and lower cost. Lap weld
joints of Zn-coated steel to austenitic stainless steel are used also in other fields such as
plate-tube joints, radiators, washing machines as well as some components in the aeronautic
field [16].
Although welding of Zn-coated steel to austenitic stainless steel is a common practice, it
presents serious problems concerning with weld zone porosity and LME cracking of
austenitic stainless steel base metal due to zinc vaporization. These welding problems have
been studied in the case of conventional metal arc welding processes [17]. It is reported that
joining gaps between the sheets to be lap welded are adjusted in order to enable the
degasification of zinc vapour.
On the other hand, laser butt and lap weld joints of both similar and dissimilar materials are
being used in many industrial applications. The fraction of laser welding in all industrial
applications is about 15-25% which varies from country to country [18-20]. The most notable
features of laser welding compared with other conventional welding processes are the high
weld quality and high welding speed. This together with its low heat input makes laser a
most hopeful candidate for thin sheet metal welding.
However, similar problems as in conventional metal-arc-welding of Zn-coated steel to
austenitic stainless steel are expected also in case of laser beam welding. Therefore, more
work is required for understanding these problems and the factors affecting them.

Welding Processes 104
This investigation has been concerned with CO2 laser welding of austenitic stainless steel in
lap joints with Zn-coated carbon steel. The focus was made on weld joint quality in terms of
weld profile, porosity in the weld zone and liquid metal emrittlement (LME) cracking of the
austenitic stainless steel base metal. The influence of type and flow rate of shielding gas, gap
between the sheets and zinc removal prior to welding was clarified. Quality of weld joints
was evaluated as a function of weld zone shape, porosity and LME cracking of austenitic
stainless steel base metal.
2.2. Experimental procedure
Commercial types of ASTM A36, 0.7 mm thick carbon steel sheet coated with 10 m zinc on
both sides and ASTM A240 Type 304L, 1mm thick stainless steel sheet were used for
dissimilar lap joints. Table 3 shows their chemical composition and mechanical properties.

Base Metal C Mn Si S P Cr Ni
YS
(N/mm
2
)
UTS
(N/mm
2
)
Elong
(%)
Zn-coated 0.04 0.35 0.26 0.01 0.02 - - 245 377 27
304L 0.04 1.70 0.35 0.01 0.03 18.2 8.5 359 558 32
Table 3. Chemical composition (wt%) and mechanical properties of used base metals
Pairs of these dissimilar steel sheets of 150x150 mm were welded with an overlap of 50 mm
and with weld bead at the middle of the overlap. Zn-coated steel sheet was upper-most and
the joint was clamped 15 mm on both sides of the weld line along its entire length.
Configurations of laser lap weld specimen are shown in Figure 12. All specimens were
ultrasonically cleaned to remove dirt and oil prior to welding.

Figure 12. Configuration of the used laser lap weld specimen.

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 105
Welding was performed using CO2 laser with a maximum output of 3 kW operated in
multi-mode. The beam was focused using a parabolic mirror with 150 mm focal length.
Laser beam welding parameters used are summarized in Table 4. Optimizing laser power,
welding speed and focal point position is of considerable importance for the weld quality in
terms of fusion zone size and profile. In order to clarify the influence of shielding gas, gap
between the sheets and pre-weld zinc removal on weld quality, laser power, welding speed,
and defocusing distance (focal point position) were optimized and kept constant at 2.5 kW, 3
m/min, and 0.1 mm below specimen surface respectively. Shielding was done using argon
or helium with flow rate of 15~30 l/min. Prescribed gap ranging from 0.025 to 0.3 mm was
introduced between the sheets along the clamped areas of the welding fixture. Pre-weld zinc
removal from the weld area was done by grinding.

P
(KW)
S
(m/min)
Dd
(mm)
Shielding gas

Type Flow rate
(1/min)
Gap between
Sheets
(mm)

Zinc removal
prior to
Welding
2.5 3 -0.1 Argon 15 ~ 30 No No
2.5 3 -0.1 Helium 15 ~ 30 No No
2.5 3 -0.1 Argon 15 0.025 ~ 0.3 No
2.5 3 -0.1 Helium 15 0.025 ~ 0.3 No
2.5 3 -0.1 Argon 15 No Yes
2.5 3 -0.1 Helium 15 No Yes
Table 4. Laser welding parameters
P: Laser power, S: Welding speed, Dd: Defocusing distance
Working distance: 10 mm at Dd = 0.0 mm, Nozzle diameter: 4mm
After welding, the specimens were subjected to non-destructive testing including visual and
dye penetrant test methods then, sectioned transverse to the welding direction. Three
sections of each seam weld were prepared for metallographic examinations using standard
technique. Quality of the dissimilar lap joints was evaluated as a function of weld profile,
porosity level in fusion zone, LME cracking in austenitic stainless steel base metal. Tensile
shear test was carried out for all laser lap welded joints and the data reported are the
average of three individual results.
2.3. Results and discussion
2.3.1. Effect of shielding gas
Examples of macrographs laser welds produced using argon and helium as a shielding gas
with a flow rate of 15 l/min are shown in Figure 13-a and b respectively. It is clear that non-
uniform weld beads with large pores were obtained in both cases. However, wide seam
width combined with an increase of the frequency of pores was obtained with argon (Figure
13-a). In other words, the number of pores was much less with using helium as a shielding

Welding Processes 106
gas (Figure 13-b). This means that shielding gas type is an essential factor to improve weld
quality since it is used to protect the molten metal against oxidation and blow the plasma
away from the beam path.
Generally, when welding zinc coated steel with stainless steel there is a very strong plasma
formation due to the low boiling point of zinc (906
o
C) and it's high vapour pressure, which
is about eight orders of magnitude greater than that of Fe [21]. The high vapourization of
zinc increases the pressure of the vapour, which is transformed to plasma in the laser beam,
to expand further into the free space above the metal surface. This will affect the absorption
and fluctuation of the plasma and in practice this is shown as increased splattering and
porosity in the weld. This plasma effect was reduced as a result of the higher ionization
potential of helium then, weld quality was improved.
With this relatively low flow rate, the process seemed to change swiftly between deep
penetration welds and vapour assisted welds. This may be explained by the presence of
zinc, which makes the process unstable due to plasma fluctuation as has reported in prior
investigation [22]. Consequently, optimized shielding gas flow rate makes the difference
between a good or poor weld.

Figure 13. Macrographs of laser welds produced using (a) argon and (b) helium as a shielding gas with
15 l/min flow rate.
With increasing shielding gas flow rate above 15 l/min, the number of pores in seem welds
was decreased and weld pool penetration was increased. Examples of macrographs of laser

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 107
welds produced using higher flow rate; 30 l/min of argon and helium as a shielding gas are
shown in Figure 14. It is obvious that smooth and homogeneous seam welds free from pores
were obtained in both cases. The increase in penetration depth obtained in this case is
consistent with the expected effect of increased plasma suppression with increased flow
rate, i.e. more of the beam was allowed to reach the work-piece. However, there appeared to
be a trade-off between plasma suppression effects and weld pool stability with increase
shielding gas flow rate.
Turning to shielding gas type, it remarkably affected weld zone profile. Low magnification
stereoscopic photographs of cross sections taken from laser welds of Figure 14-a and b are
shown in Figure 15-a and b respectively. The use of helium has resulted in complete
penetration with higher depth/width ratio and a slight taper configuration which means
minimum fusion zone size (Figure 15-b) in comparison with that obtained using argon
(Figure 15-a). In other words, helium has a more favorable effect on the molten metal than
argon at optimized flow rates, which make the welds more homogeneous and free from
pores. A flow rate of 22 l/min for helium was found to be satisfactory in comparison with 30
l/min for argon. These results do conform with prior results of other investigators where
they have shown that the weld defects, due to the vapourization of zinc, can be reduced by
optimizing shielding gas parameters [23-25].


Figure 14. Macrographs of laser welds made using (a) argon and (b) helium as a shielding gas with
30 l/min flow rate.

Welding Processes 108

Figure 15. Low magnification stereoscopic photographs of cross sections taken from laser welds of
Figure 14
On the other hand, it is found that both type and flow rate of shielding gas have no effect on
LME cracking of austenitic stainless steel. Figure 16 shows a typical example of optical
micrograph of a cross section taken from lap weld produced using helium with its optimum
flow rate; 30 l/min. In spite of obtaining homogeneous, sound, complete penetration and
acceptable weld profile with such high flow rate, the noticeable feature is the formation of
severe cracking at the stainless steel base metal. These cracking were extended for a distance
of about 0.7 mm around both sides of lap weld joints and propagated on grain boundaries.
This type of cracking is typical LME cracking of the austenitic stainless steel which occurs
above 750
o
C when it is exposed to molten zinc and tensile stresses. Molten zinc can be
produced by the heat of welding and tensile stresses can be generated from the heating and
cooling cycles during welding. This cracking type is characterized by extremely rapid crack
propagation perpendicular to the applied stress [26]. It should be mentioned also that
similar results were obtained for lap weld joints produced using argon gas shielding.

Figure 16. Typical example of optical micrograph of a cross section taken from laser lap weld produced
using helium as a shielding gas with its optimum flow rate; 30 l/min.

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 109
2.3.2. Effect of gap between the sheets
In the above section, lap welds were made with a good contact, i.e. without a gap between
the sheets. In order to clarify its effect on weld joint quality, prescript gap was introduced
between the sheets and shielding was done using either argon or helium with its lower flow
rate (15 1/min) which resulted in weld porosity in the previous experiments.
Typical examples of cross sections taken from laser lap welds produced using 0.0 and 0.025
mm gap between the sheets in case of helium shielding are shown in Figure 17-a and b
respectively. As has been explained in the previous section, no-gap welds with such low
shielding gas flow rate showed unacceptable levels of porosity with varying amounts of top
surface undercutting and center-line humping of the weld bead (Figure 17-a). The zinc
vapourization from the underside of the joint in the case of complete weld penetration is not
sufficient to prevent porosity formation. Introducing a small gap distance between the
sheets with same welding conditions resulted in a sound weld where porosity was not
observed (Figure 17-b). Also, weld profile was remarkably improved where a smooth
curved and symmetrical fusion zone interface was obtained.
In other words, acceptable quality for laser lap welds concerning soundness and profile
could not be obtained with low shielding gas flow rate and without a gap between the
sheets. Once the heat input was sufficient to permit melting through the top sheet, there was
explosive ejection of molten weld metal due to vapourization of the zinc layers at the Zn-
coated steel sheet-to-stainless steel sheet interface. This resulted in extensive weld metal
porosity or complete expulsion of the weld metal in the case of no-gap weld leading to
undercutting the top steel sheet, as shown in Figure 17-a. This is in a good agreement with
results of previous investigations [27, 28].

Figure 17. Typical examples of cross sections taken from laser lap welds produced using 15 l/min
helium and different gap distances. (a) Gap: 0.0mm, (b) Gap: 0.025mm.
Generally, no porosity was found in any of the welds made with introducing gap between
the sheets. However, welds made by gap larger than 0.05 mm showed unacceptable weld
profile where weld depth /width ratio decreases sharply and the weld geometry begins to

Welding Processes 110
deteriorate. Photographs of laser weld cross sections produced using 0.1 and 0.3 mm gaps
are shown in Figure 18-a and b respectively. A gap distance of 0.1 mm gave a concave top
surface, with a relatively low depth/ width ratio (Figure 18-a). The tendency of the molten
pool to collapse increased significantly with increasing the gap to higher value. This resulted
in remarkable undercut in the welds and excessive drop-through of the weld metal into the
gap (Figure 18-b).

Figure 18. Photographs of cross sections taken from laser lap welds produced using (a) 0.1mm and (b)
0.3mm gap between the sheets.
Generally, there are two mechanisms in laser welding, one is heat conduction under low
energy density and the other is deep penetration (keyhole effect) under high energy density.
In these experiments, the laser power density when laser beam touched the surface of the
top sheet was high enough to melt the metal rapidly and formed the deep penetration. The
power density greatly decreased when approached the bottom plate particularly in case of
large air gap between both plates that obstructed heat transfer. At this time, the heat transfer
was mainly by means of conduction, which could be proved by the weld shape and
penetration depth. Based on the weld shape, the fusion lines at both sides of welds were
approximately parallel under deep penetration welding conditions, while the fusion line
was half circle under heat conduction welding. Consequently, under such experimental
conditions, the welding mechanism of lap joints was combination of deep penetration and
heat conduction.
Although introducing a small gap between these dissimilar material sheets has resulted in
avoiding porosity in weld zone, it has no effect on zinc induced LME cracking in austenitic
stainless steel base metal. Figure 19 shows typical example of an optical micrograph of a
cross section taken from lap welded joint produced using helium with 15 1/min flow rate
and 0.05 mm gap. It is noticed that a sound and uniform weld seam was obtained. However,
the most important notice is the formation of LME cracking on grain boundaries of stainless
steel base metal as has been explained in the previous section. Cracking were extended for a
distance of about 0.5 mm around both sides of lap weld joints. These results of laser welding
do conform to other research work concerned with arc welding processes [17].

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 111
Generally, the influence of gap between the sheets on weld geometry and quality can be
explained using schematic illustrations shown in Figure 20. Since the zinc vapour has no
enough access escape route with zero gap, both pores weld with unacceptable profile and
LME cracking in stainless steel will be obtained (Figure 20-a). Introducing a small gap
between the sheets give the zinc vapour an alternative escape route during welding then,
sound and acceptable weld joint will be produced. It should be mentioned that this gap
should be limited to an optimum value as has been reported by other researchers [22].
However, the problem is still concerned with LME cracking in austenitic stainless steel base
metal around both sides of the joint since it can not be prevented by these measures (Figure
20-b).
2.3.3. Effect of zinc removal prior to welding
The results of the previous two sections confirmed that LME cracking of austenitic stainless
steel in laser lap joint with Zn-coated steel is attributed mainly to molten zinc resulted from
welding heat. Consequently, this section is concerned with studying this type of cracking as
a function of zinc removal prior to welding. In this respect, the effect of both one and two
sides grinding of the weld area of Zn-coated sheets was clarified using welding parameters
previously resulted in weld zone porosity.

Figure 19. Typical example of optical micrograph of a cross section of lap welded joint produced using
0.05mm gap and 15 l/min helium.
Photographs of laser welds produced without gap between the sheets and with 15 1/min of
helium shielding after zinc removal by grinding of weld area from one and two sides are
shown in Figure 21-a and b respectively. It can be noticed that removing zinc coating from
only one side of Zn-coated sheet was not effective to obtain sound welds since porosity were
observed in weld zone (Figure 21-a). In the case of two sides grinding before welding,
molten zinc was avoided due to removing of zinc coating then, molten metal was not ejected
and this in turn could result in a sound and uniform weld seam (Figure 21-b).

Welding Processes 112

Figure 20. Schematic illustrations showing the effect of introducing a gap between the sheets being
welded on weld zone profile, porosity and LME cracking.

Figure 21. Photographs of laser welds produced using 15 l/min helium shielding and zero gap between
the sheets after zinc removal from (a) one side and (b) two sides of the weld area of Zn-coated sheet.

Figure 22. Effect of two sides grinding of weld area of Zinc-coated sheet on LME cracking
susceptibility.

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 113
In addition, the most important finding in the case of two sides grinding is the
disappearance of LME cracking in austenitic stainless steel as shown in Figure 22 that could
not be attained in the above two sections. This is due to the complete removal of zinc
coatings from the weld area prior to welding.

Figure 23. Tensile shear strength of laser lap welded joints as a function of welding conditions used.
Welding conditions I, II and III are the optimum conditions used in section 3.3.1, 3.3.2 and 3.3.3,
respectively.
Results of tensile shear test of laser lap joints as a function of welding conditions used are
shown in Figure 23. Tensile shear strength of joints produced with zinc removal from both
sides prior to welding was considerably higher than that of all other joints produced with
and without gap regardless of shielding gas type and flow rate. This is attributed to the
absence of LME cracking in case of two sides grinding.
Recently, serious industrial incidents of zinc-induced LME cracking in austenitic stainless
steel have been reported [29, 30]. The potential for cracking during field welding is certainly
greater than the cracking potential in these test specimens. This is due to higher tensile
stresses in the case of field welding. Therefore, the removal of galvanized zinc coating prior
to welding should be done properly to avoid contamination of austenitic stainless steel with
any molten zinc during welding.
3. Conclusion
For CO2 laser butt welding of similar butt joints of austenitic stainless steels, the
following conclusions can be drawn.
- The penetration depth increased with the increase in laser power. However, laser
power has a less effect on weld profile.
- Unlike laser power, welding speed has a pronounced effect on size and shape of the
fusion zone. Increase in welding speed resulted in an increase in weld depth/width
ratio and hence a decrease in the fusion zone size.

Welding Processes 114
- Minimizing heat input and optimizing energy density through optimizing laser power,
welding speed, and defocusing distance is of considerable importance for the weld
quality in terms of fusion zone size and profile. Helium is more effective than argon as a
shielding gas to obtain acceptable weld profile.
- Fusion zone composition was insensitive to change in heat input. However, increase in
welding speed and/or decrease in laser power resulted in a finer solidification structure
due to low heat input. A dominant austenitic structure with no solidification cracking
was obtained for all welds. This could be associated with primary ferrite or mixed mode
solidification based on Suutala and Lippold diagrams.
- Mechanical properties, tensile, hardness and bending at room temperature, were not
significantly affected by heat input.
For CO2 laser welding of austenitic stainless steel in lap joints with Zn-coated carbon
steel, the following conclusions can be drawn.
- One way to produce sound and uniform laser lap welds of Zn-coated steel with
austenitic stainless steel without gap between the sheets is the optimizing shielding
conditions. This is of considerable importance for avoiding plasma or preventing
porosity and obtaining full penetration without deteriorating the surface quality of the
weld. Helium shielding produced noticeably deeper welds while argon exhibited the
smoothest top surface. A flow rate of 22 1/min was found to be satisfactory in the case
of helium in comparison with 30 1/min for argon.
- The other way to produce sound and homogeneous laser lap welds of these dissimilar
materials is the introducing of a small gap (0.025~0.05 mm) between the sheets.
Maintaining such gap between the sheets give the zinc vapour an alternative escape
route. Smaller gap resulted in pores weld and random instabilities in the weld bead
surface while larger gap showed unacceptable levels of drop-through of the weld metal
between the sheets.
- To preclude both weld porosity and cracking of the stainless steel by molten zinc in
making attachments of Zn-coated steel to 300 series austenitic stainless equipment
which in turn will improve tensile shear strength, the choice seems to be very clear that
is the zinc coating must be scrupulously removed from the joint area prior to welding.
Recently, new generation of lasers, such as fiber and disk lasers is receiving great
attention due to its high efficiency, high power and high beam quality, which can produce
an ultra-high peak power density of MW/mm2 levels corresponding to a focused electron
beam. These features and advantages of fiber and disc lasers are of considerable importance
for deep penetration and high speed welding of austenitic stainless steels with thick
sections, which are used in some critical applications such as nuclear power plants. It has
been reported that such new generation of lasers is promising to be among the desirable
heat sources for deep-penetration high speed welding of thick-section austenitic stainless
steels [31-34].
In this regard, multi-passes narrow-gap welding of 50mm thick 316L plates has been
investigated using 8 kW disk laser where the effect of welding conditions on the weld bead

Laser Beam Welding of Austenitic Stainless Steels Similar Butt and Dissimilar Lap Joints 115
geometry and welding defects was studied. It is reported that butt joint of 50 mm thick
plates with narrow gap could be performed with eight-layers welding at laser power of 6
kW and welding speed of 0.4 m/min. In order to reduce the weld passes further, gas jet
assisted laser welding was tried to weld thick 316L plates with a 10 kW fiber laser. The
result showed that butt-joint welding of 40 mm plates without filler wire could be carried
out at 0.3 m/min welding speed with no porosity or other welding defects. As for 50 mm
thick plate, a good weld bead could be obtained with bead-on-plate welding from both sides
at 0.2 m/min welding speed [35].
Author details
Abdel-Monem El-Batahgy
Manufacturing Technology Department, Central Metallurgical R & D Institute, Cairo, Egypt
Acknowledgement
The author would like to thank Laser-X Company Ltd., Japan for conducting CO2 laser
beam welding experiments.
4. References
[1] Zacharia, T.; David, S. A.; Vitek, J. M. & Debroy, T. (1989). Metall. Trans. 20 A, 1125.
[2] Suutala, N. (1983). Metall. Trans. 14 A, 191.
[3] Klimpel, A. & Lisiecki, A. (2007). Laser Welding of Butt Joints of Austenitic Stainless
Steel AISI 321. Journal of Achievements in Materials and Manufacturing Engineering, 25, 1.
[4] Curcio, F.; et al. (2006). Welding of Different Materials by Diode Laser. Journal of
Materials Processing and Technology, 175, 83-89.
[5] Zacharia, T.; David, S. A.; Vitek, J. M. & Debroy, T. (1989). Welding Journal, 68, 12.
[6] David, S. A.; Vitek, J. M. & Hebble, T. L. (1987). Welding Journal. 66, 289.
[7] Schaeffler, A. L. (1949). Metal Progr. 56, 680.
[8] Kujanpaa, V.; Suutala, N.; Takalo, T. & Moisio, T. (1979). Welding Research Int. 9, 55.
[9] Vitek, J. M. & David, S.A. (1982). ASM Conference Proceedings, Trends in Welding
Research in the United States (ASM, Metals Park, Ohio,) pp.243-258.
[10] Lippold, J. C. (1985). Welding Journal, 64, 127.
[11] Kotecki, D. & Siewert, T. A. (1992). Welding Journal 71, 171.
[12] Lippold, J. C. (1994), Welding Journal. 73, 129.
[13] Dilthey, U. & Risch, A. (2001). Laser Welding of Stainless Steels and Stainless Low-
Alloy Material Combinations, Welding in the World, 36, 67-71.
[14] Elmer, J. W.; Alien, S. M. & Eagar, T. W. (1990). Recent Trends in Science and Technology,
eds. David S. A. & Vitek J. M. (ASM International, Materials Park, Ohio, pp. 165-170.
[15] Brooks, J. A. & Thompson, A. W. (1991). Inst. Met. Reviews, 36, 16.
[16] Metzbower, E. A. (1991). Laser Welding, Naval Engineers Journal, Vol. 8, p. 41-49

Welding Processes 116
[17] Bruscato, R. M. (1992). Liquid Metal Embrittlement of Austenitic Stainless Steel When
Welded to Galvanized Steel. Welding Journal, 71(12), p.455s-459s.
[18] Belforte, D. A. (1990). Annual technology, Industry, and Market Review and Outlook.
The Industrial Laser Annual Handbook, Pennwell Books Laser Focus, Tulsa, Oklahoma
USA, SPIE 1241, p. xvi-xxxi.
[19] Roessler, D. M. (1989). Laser Processing-Global Overview and Future Trends.
Proceedings of the 21st International Symposium on Automotive Technology and Automation,
Vol. 1, Wiesbaden, Germany, November 1989, p. 494-516.
[20] Marinono, G.; et al. (1989). Technical and Economic Comparison of Laser Technology
with The Conventional Technologies For Welding. Proceedings of the 6th Int. Conf. on
Laser in Manufacturing, IFS, May 1989, ISBN 1-85423-047-6, p.105.
[21] Beyer, E. & Gasser, E. (1987). Plasma Fluctuations in Laser Welding With CW CO2-
Lasers. Proc. of The 6th Int. Cong. on Appl. Lasers and Electro-Optics, ICALEO 87, San
Diego, California, November 1987, p. 17-23.
[22] Bagger, C.; Miyamoto, I.; Olsen, F. & Maruo, H. (1992). Process Behaviour During High
Power CO2 Laser Welding of Zinc Coated Steel. Proceedings of LAMP, Nagaoka, Japan,
June 1992.
[23] Akhter, R.; Steen, W. M. & Cruciani, D. (1988). Laser Welding of Zinc Coated Steel. Proc.
of The 5th Int, Cong. on Lasers in Manufacturing, LIM 5, Stuttgart, West Germany,
September 1988, p. 195-206.
[24] Heyden, J.; Nilsson, K. & Magnusson, C. (1989). Laser Welding of Zinc Coated Steel,
Proc. of The 6th Int, Conf. on Lasers in Manufacturing, LIM 6, May 1989, p. 93-104.
[25] Graham, M. P.; Hirak, D. M.; Kerr, H. W. & Weckman, D. C. (1994). Nd-YAG Laser
Welding of Coated Sheet Steel. Journal of Laser Applications, Vol. 6, p. 212-222.
[26] Stoloff, N. S. (1977). Recent Development in Liquid Metal Embrittlement. Proceedings
Conference on Environment Sensitive Mechanical Behavior, AIME, Chicago, 1977.
[27] Nicholas, M. G. & Old, C. F. (1979). Review of Liquid Metal Embrittlement. Journal of
Material Science, Vol. 14, p. 1-18.
[28] Tzeng, Y. (2006). Gap-free lap welding of zinc-coated steel using pulsed CO2 laser.
International Journal of Advanced Manufacturing and Technology, 29, 287-295.
[29] Johnson, J. M.; Berry, M. R. & Gutzeit, J. (1982). Zinc Embrittlement of Stainless Steel
welds. Proceedings of AIME Meeting on Embrittlement by Liquid and Solid Metals, St. Louis,
MO, 1982.
[30] Shinohara, T. & Matsumoto, K. (1982). Welding Cracks of Zn-contaminated Stainless
Steel Pipe. Corrosion Science, 22(8), p.723-737.
[31] Thomy, C.; Seefeld, T. & Vollertsen, F. (2005). Proceedings of the Third International
WLT-Conference on Lasers in Manufacturing, Munich, Germany, pp.27-32.
[32] Verhaeghe, G. & Hilton, P. (2005). Proceedings of ICALEO, Miami, USA, pp.264-271.
[33] Liu, S.; Kutsuna, M. & Xu, G. (2006). Proceedings of ICALEO, Scottsdale, USA, pp.562-568.
[34] Reem, S. (2006). Proceedings of ICALEO, Scottsdale, USA, pp.586-594.
[35] Zhang, X.; Ashida, E.; Tarasawa, S.; Anma, Y.; Okada, M.; Katayama, S. & Mizutani, M.
(2011). Welding of thick stainless steel plates up to 50 mm with high brightness lasers.
Journal of Laser Applications, 23, 022002 (2011); http://dx.doi.org/10.2351/1.3567961
Chapter 6




2012 Ma et al., licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Mitigating Zinc Vapor Induced Weld Defects in
Laser Welding of Galvanized High-Strength Steel
by Using Different Supplementary Means
Junjie Ma, Fanrong Kong, Blair Carlson and Radovan Kovacevic
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/53562
1. Introduction
Laser beam welding is a process where a focused laser beam is used as a moving heat source
to join pieces of metal. The focused laser beam has a high power density that allows high
speed welding, a deep penetration and a narrow heat affect zone (HAZ). There are two
distinct types of laser welding modes: a conduction welding mode and a keyhole welding
mode. When the laser beam intensity reaches 10
9
W/m
2
, the molten pool starts to evaporate.
As the laser intensity increases above 10
10
W/m
2
, the recoil pressure of the metal vapor
pushes the molten metal downward and aside and a deep capillary called the keyhole is
generated (Dawes, 1992). In the keyhole mode welding process, the keyhole maintains open
due to the dynamic balance between the liquid metal surface tension and the pressure of the
metal vapor and laser-induced plasma (Bakowski et al., 1984). When the laser radiates on
the wall of the keyhole, the laser reflects multiple times on the wall of the keyhole. The laser
beam energy is absorbed by Fresnel absorption directly by the walls of the keyhole, and a
fraction of the laser energy is absorbed during each reflection (Dowden, 2009). Due to the
multiple reflections of the laser beam, the keyhole behaves like an optical black body,
making the keyhole mode welding process a highly energy efficient one (Steen, 2003).
Lap joint is the most common type of joint in the automotive assembly application; the
traditional car body assembly method in a lap joint configuration uses resistance spot
welding. However, the heavy and big spot guns limit the flexibility and accessibility of the
welding process (Park et al., 2010); moreover, the localized joints are not particularly strong
compared to those acquired by laser welding. On the other hand, the laser welding provides
several benefits including a high scanning speed, high strength and low distortion of the
joints, and the flexible implementation of the system for the automotive industry. Because of

Welding Processes 118
these advantages, laser welding shows immense potential over the conventional resistance
spot welding and has been widely used in the automotive industry in the fabrication of
different auto bodies parts (Forrest et al., 2004).
In order to reduce the weight of the vehicles and improve fuel efficiency and safety, the
development of lightweight, and high-strength vehicles has prompted an increased use of
advanced high strength steels (AHSS) in the automotive industry. These new steel grades
include dual phase (DP) steels, transformation-induced plasticity (TRIP) steels, high hole
expansion (HHE) steels, complex-phase (CP) steels, martensitic steels (MS), and twining
induced plasticity (TWIP) steels (UltraLight Steel Auto Body- Advanced Vehicle Concepts,
2001). Additionally, these steels are galvanized in order to improve the surface corrosion
resistance for automotive parts. However, it is still a great challenge to laser weld of
galvanized steels in a zero-gap lap joint configuration. When laser welding of galvanized
steels in a zero-gap lap-joint configuration, the zinc coating at the contact interface will
vaporize; due to the lower boiling point (906 C) of zinc as compared to the melting
temperature of steel (above 1500 C), the highly pressured zinc vapor expels the liquid metal
out of the weld pool, resulting in blowholes and pores which dramatically decrease the
mechanical properties of the weld (Akhter et al., 1988) (Fig. 1).

Figure 1. (a) The schematic view of the laser welding of galvanized steel in a zero gap lap-joint
configuration and (b) the acquired weld bead with pores
2. Review of the laser welding of galvanized steels in a lap joint
configuration
Over the past several decades, industry and academic researchers have been seeking new
technologies that will successfully join galvanized steels in a lap joint configuration. Many
techniques were proposed, and some are listed in Table 1.
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 119
Methods Schematic Technical details Drawbacks
Elongating the laser beam
(Fabbro et al., 2006) or using
low power / low speed laser
welding (Ma et al., 2012)
The zinc vapor was
degassed through the
keyhole or the
enlarged molten pool
during welding
Low processing
speed or unstable
weld qualities that
limit the
application in
production
Using pulsed laser welding
(Heydon et al., 1989; Kennedy
et al., 1989; Norris et al., 1992;
Tzeng, 1999; Tzeng, 2006)

Zinc vapor was
mitigated in the
pulsed laser welding
and effectively
exhausted through a
stable keyhole
Using various shielding gas
combinations (Berlinger, 1987;
Akhter et al., 1990; Ream,
1991; Mitsubishi Co., 1993;
Chung et al., 1999; Briand et
al., 2008; Yang et al., 2011)

Suppressed the
formation of the laser
induced zinc plasma or
interacted of the zinc
vapor with the
shielding gas during
welding
Pre-placing a thin metal sheet
or powder along the
centerline of the weld seam
(Dasgupta et al., 2000; Li et al.,
2007)
The zinc reacted
chemically with the
added metal before
the steel started to
melt
Difficulties will be
implemented in
production
Applying an appropriate
spacer at the faying surfaces
(Akhter et al., 1988; Imhoff et
al., 1988)
The generated zinc
vapor vented out
through the gap
The additional pre-
welding procedure
increases the
production cost
Using a laser to create humps
on the bottom sheet to create a
gap at the faying surfaces (Gu
et al., 2011)
The generated zinc
vapor vented out
through the gap
Creating vent holes on the
bottom steel sheet (Chen et al.,
2009)
The generated zinc
vapor vented out
through the vent
holes
Adding a second laser heat
source or splitting the laser
beam into two laser beams in
order to weld galvanized steel
(Loredo et al., 2002; Xie et al.,
2001)
The leading laser
melted the zinc
coating at the
interface
Complex
equipment that
would be difficult
to implement in
the production
environment.

Welding Processes 120
Using a leading beam to cut a
slot along the joint line (Iqbal
et al., 2010; Milberg et al.,
2009)
The leading laser cut
the slot through
which the zinc vapor
was vented out
Moreover, a
specific offset is
needed between
these two heat
sources that could
limit the
application of this
welding procedure
Applying gas tungsten arc
welding (GTAW) as an
auxiliary preheating heat
source (Gu et al., 2001; Kim et
al., 2008; Yang et al., 2009)
The preheating
increased the
absorption of the
laser beam that
contributes to a
formation of a stable
keyhole through
which the zinc vapor
is evacuated
Table 1. The alternatives for the welding of galvanized steels in a lap joint configuration
As mentioned above, all of these techniques have some drawbacks that limit their
application in the industry. So far, there is no report on a cost-effective, efficient, and easy-
to-apply welding procedure that is capable of joining galvanized high-strength steels in a
zero-gap lap joint configuration without the material addition or without the assistance of a
second heat source. In order to avoid highly pressurized zinc vapor caused weld defects like
spatters and blowholes, either the zinc coating should be removed before the steel melts or
the vaporized zinc needs to be vented out properly during the welding process. The most
direct way of accomplishing this is to mitigate the presence of high pressure zinc vapor
during the welding process. In order to achieve this goal several techniques are proposed
which will be described in the following sections of this chapter.
3. Low power / low speed laser welding of galvanized steels in a zero-gap
lap joint configuration
3.1. Experimental procedure
Ribic et al. (2009) concluded that the generated zinc vapor escapes from the weld pool if the
solidification time is longer. A lower welding speed will generate an enlarged weld pool
that will require a longer solidification time. An experimental work is presented to show
that the effect of zinc vapor on the quality of a weld in a zero-gap lap joint configuration
may be successfully mitigated. A fiber laser of 4 kW in power with a focused spot of 0.6 mm
in diameter was used as the welding heat source, and a 6-axis high precision robot was used
to implement the welding procedure of galvanized steels (see Fig. 2). Pure argon with a flow
rate of 30 standard cubic feet per hour (SCFH) was employed as side shielding gas to
suppress the laser-induced plasma and to protect the molten material against corrosion. The
base material used in this work was galvanized high strength dual phase (DP) steel DP980,
whose nominal chemical composition is listed in Table 2 (Burns, 2009). The coupons of
galvanized DP980 steel sheets were 1.2 mm and 1.6 mm in thickness, with a zinc coating
weight of about 60 g/m
2
.
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 121

Figure 2. Experimental setup for low power / low speed laser welding

C Si Mn Cr Mo Al Cu Ni Fe
0.15 0.31 1.5 0.02 0.01 0.05 0.02 0.01 Balance
Table 2. Chemical composition of galvanized DP980 steel, wt% (Burns, 2009)
3.2. Experimental results for low power / low speed laser welding
The top and bottom views of the weld obtained under a laser power of 2.0 kW and a
welding speed of 5 mm/s show an acceptable weld surface quality (see in Fig. 3). A high-
speed CCD camera with a frame rate of 4000 fps combined with a green laser with a band
pass filter wavelength of 532 nm as the illumination source were used for real time
monitoring of the dynamic behavior of the molten pool under different laser welding
conditions. The weld pool formed under a relatively low welding speed was larger and
relatively stable (see Fig.4a). On the other hand, the molten pool acquired under a higher
welding speed shows sever fluctuation, and the high pressured zinc vapor generated at the
faying surface jetted into the molten pool causing blowholes (see Fig. 4b). According to
Ribics work, an enlarged weld pool has a longer solidification time which obviously
decreased the probability that the zinc vapor would be trapped in the molten pool under a
relatively low welding speed (around 5 mm/s), and a visually acceptable weld quality could
be acquired. However, if the welding speed is exceedingly low, the sagging may be
generated, which also reduces joint strength. Fig. 5 shows the tensile shear test results of the
joints acquired under different welding speeds. A higher failure load was acquired under a
lower welding speed. The trapped zinc vapor may result in pores inside the joints which
could decrease the failure load with respect to the joints acquired under the same welding
conditions but without zinc at the faying surface (see Fig. 5). Although an acceptable quality

Welding Processes 122
of joints could be achieved by this low power / low welding speed procedure, this
procedure is not accepted by the industry because of a low productivity.

Figure 3. Top and bottom views of the weld obtained under a laser power of 2.0 kW and welding speed
of 5 mm/s

Figure 4. The dynamic behavior of the molten pool acquired under different welding parameters: (a)
laser power of 1.5 kW, welding speed of 5 mm/s, and (b) laser power of 4.0 kW, welding speed of 30
mm/s at different time steps during the welding process
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 123

Figure 5. The failure load for tensile shear testing of the DP980 lap joints acquired under different
welding speeds and with the laser power of 2.0 kW
4. Two-pass laser welding of galvanized steels in a zero-gap lap joint
configuration
4.1. Experimental procedure
In order to improve the production efficiency, the automotive industry requires a welding
technique that can join of overlapped galvanized high strength steels successfully under a
higher working speed. As discussed previously, if the zinc coating is removed before the
steel starts to melt, a much higher welding speed can be achieved. Therefore, a two-pass
laser welding process that is capable of successfully joining galvanized steel sheets in a zero-
gap lap joint configuration is presented. Fig. 6 shows the schematic view of the two-pass
laser welding process. The defocused laser beam shown in Fig. 6a is used to preheat the two
overlapped sheets during the preheating pass. Only when the width of the area where the
zinc coating is vaporized by preheating is larger than the distance between the zinc boiling
isotherms (906 C), a sound weld could be acquired (see Fig. 6c). The laser power was set at
its maximum value of 4.0 kW in order to allow a higher scanning speed. The laser welding
speed was set at 60 mm/s in order to acquire a partial penetrated joint that was determined
by the preliminary executed experimental trails. The preheating parameters, like the
defocused position of the laser beam and the laser scanning speed, are critical in achieving a
final weld quality. As shown in Table 3, three levels of defocused off-set distance (26 mm, 44
mm, and 62 mm which corresponds to the defocused diameters of laser beam of about 3
mm, 5 mm, and 7mm, respectively) and four levels of scanning speeds (20 mm/s, 30 mm/s,
40 mm/s, and 50 mm/s) were chosen to optimize the preheating procedure.

Welding Processes 124

Figure 6. The schematic view of the two-pass welding process: (a) laser preheating, (b) laser welding,
and (c) geometrically defined width of zinc coating treated during preheating and welding

Process parameters Experiment series A Experiment series B Experiment series C
Scanning speed (mm/s) 20 30 40 50 20 30 40 50 20 30 40 50
Defocused off-set distance
(mm)
26 44 62
(Laser preheating and welding power: 4.0 kW, laser welding speed: 60 mm/s)
Table 3. The preheating parameters
4.2. Experimental results for two-pass laser welding
As the defocused off-set distance increased, the defocused laser beam spot became larger,
and the laser energy distribution was dispersed. A defocused off-set distance combined with
a lower scanning speed generated too much energy that penetrated the top sheet resulting
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 125
in spatters and permanent defects which could not be mitigated by the following laser
welding pass (see Fig. 7a). A longer defocused off-set distance combined with a higher
scanning speed could not vaporize a sufficient amount of zinc coating; the remaining zinc
coating at the contact interface caused weld defects (see Fig. 7b). Thus, only for the
optimized laser defocused off-set distance and the scanning speed, will a reasonable width
of the zinc coating be vaporized (see Fig. 7c). Fig. 8 shows the experimental results for the
selected preheating parameters (shown in Table. 3). The optimized preheating parameters
that allowed a sound weld are shown in Fig. 8, area A.

Figure 7. The schematic view of the preheating process: (a) molten pool penetrates the interface, (b)
narrow vaporized zinc coating area, and (c) optimized width of the vaporized zinc coating area

Figure 8. Experimentally determined combinations of defocused off-set distance and scanning speed
that result in a good weld quality

Welding Processes 126
Figs. 9 and 10 show the preheated interfaces of the coupons and the cross-sections
corresponding to the different locations marked on the preheated interfaces. The zinc
coatings far from the preheated zones at the top and bottom sheets are not affected (see Figs.
9b and 10b); the zinc coatings are melted and deformed at the edges of the preheated zone
(see Figs. 9c and 10c); and the zinc coatings are vaporized at the center of the preheated zone
(see Figs. 9d and 10d).

Figure 9. Preheated surface of the top steel sheet obtained with a laser power of 4.0 kW, a 30 mm/s
scanning speed, and a 62 mm defocused laser beam off-set distance and the corresponding cross-section
at different locations
During the laser preheating pass, the defocused laser beam burns the zinc at the top surface,
and melts and partially vaporizes the zinc coatings at the interface of the two overlapped
steel sheets, and improves the absorption of the laser beam which results in the formation of
a stable keyhole through which any zinc vapor formed at the interface will be vented out
(Yang et al., 2009). Fig. 11 shows the top and bottom views of the weld obtained under
preheating and welding with a laser power of 4.0 kW, a defocused off-set distance of 62 mm,
a scanning speed of 30 mm/s, and a welding speed of 60 mm/s. Fig. 12 shows the weld cross-
section of the weld shown in Fig. 11.
The tensile shear test was carried out in order to determine the mechanical strength of the
welded joints obtained by the two-pass laser welding procedure. The experimental results
demonstrated that the two-pass welded joints were broken in the HAZ of the bottom steel
sheet. One of the tensile shear results is shown in Fig. 13. The tensile shear test for the
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 127
welded coupons without a zinc coating at the interface was also performed. In order to use
the data as a reference, the welded coupons without a zinc coating at the interface had a
average failure load value of 5295.88 N which was lower than that of the two-pass welded
coupons (6127.58 N). The reason for this difference in results is explained by the fact that the
preheating process increased the laser beam absorption of the coupons, which contributed
to a stronger (wider) partially penetrated weld joint.

Figure 10. Preheated surface of the bottom steel sheet obtained with a laser power of 4.0 kW, a 30 mm/s
scanning speed, and a 62 mm defocused laser beam off-set distance and the corresponding cross-section
at different locations

Figure 11. Top and bottom views of weld obtained under a scanning speed of 30 mm/s and a defocused
off-set distance of 62 mm (the preheating and welding laser power is 4.0 kW; the welding speed is 60
mm/s)

Welding Processes 128

Figure 12. Cross-sectional view of the weld obtained under a scanning speed of 30 mm/s and a
defocused off-set distance of 62 mm (the preheating and welding laser power is 4.0 kW; the welding
speed is 60 mm/s)

Figure 13. Tensile shear test results of the weld joint obtained under a scanning speed of 30 mm/s, and
a defocused laser beam off-set distance of 44 mm (the preheating and welding laser power is 4.0 kW; the
welding speed is 60 mm/s)
5. Laser welding of galvanized steels in a lap joint configuration with a
pressure wheel
5.1. Experimental procedure
Based on the experimental study performed, it was found that the stability of the laser
welding process was sensitive to the clamping conditions. A relatively loose clamp
condition resulted in a better weld than a very tight clamp condition. The gap ahead of the
weld pool is the key to performing the laser welding of galvanized steel in a lap joint
configuration successfully. Moeckel et al. (2003) developed a device for controlling the gap
at the faying surface of the overlapped galvanized steel sheets in order to degas the
generated zinc vapor during the welding process. The Fraunhofer Institute developed a
pressure wheel system which could control the roller clamping force that allows for the
controlling of the gap at the faying surface (Fraunhofer Institute website). In order to
achieve an over lapped galvanized steel joint with a single laser beam without a pre- and/or
post-weld process, a force-controllable pressure wheel (ZM YW50 PW P300 II) is used to
control the gap near the laser focused spot during the laser welding. The laser welding of
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 129
galvanized steels for a lap joint configuration with a pressure wheel control system is shown
in Fig. 14. Fig. 15a shows the close-up of the pressure wheel set-up. The laser head is set-up
under a 30 decline with respect to the pressure wheel, and Fig. 15b shows the pressure
wheel controller.

Figure 14. Laser welding of galvanized steel for a lap joint configuration with a pressure wheel control
system

Figure 15. (a) The close-up of pressure wheel set-up and (b) the pressure wheel controller
5.2. Experimental results for laser welding with a pressure wheel
The feasibility of welding galvanized steel sheets in a lap joint configuration by controlling
the pressure wheel force during the fiber laser welding process is discussed. Fig. 16 shows

Welding Processes 130
the welds obtained by various levels of pressure wheel force under a laser power of 4.0 kW
and a welding speed of 50 mm/s. The corresponding weld cross-sections are shown in Fig.
17. The cross-sections of the welds show that the weld beads are under the angle because the
laser head is set-up with an angle of 30 with respect to the pressure wheel (see Fig. 15a). A
sound weld was obtained by using a single laser beam with a force-controllable pressure
wheel under the optimized force. As shown in Figs. 16 and 17, with an increased pressure
wheel force, the weld quality decreased, and lots of spatters and blowholes were generated
(see Fig. 16d). An increased pressure wheel force larger than 12N resulted in a decreased
gap between the overlapped sheets near the laser focused area; the gap became too narrow
to evacuate the generated high pressured zinc vapor. The jet of high pressured zinc vapor
generated spatters and blowholes during the welding process.

Figure 16. Top and bottom views of the welds acquired by various pressure wheel forces under a laser
power of 4.0 kW and a welding speed of 50 mm/s
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 131

Figure 17. Cross-sectional views of the welds acquired by various pressure wheel forces under a laser
power of 4.0 kW and a welding speed of 50 mm/s
Kong et al. (2012) reported that there is a correlation between the optical emission of the
plasma and zinc vapor induced welding defects in the laser welding of galvanized steel for
an overlapped joint configuration. Therefore, the spectroscopy was used to on-line monitor
the laser welding of galvanized steel with a pressure wheel for an overlap joint
configuration. The set-up to monitor the optical emission of the plasma in laser welding is
shown in Fig. 18.
The emission line intensities detected from plasma during the laser welding of galvanized
steel under various pressure wheel forces are shown in Fig. 19. The intensities of the
emission lines above the weld pool were much lower when the pressure wheel force is
larger than 12 N; while, the intensities of the emission lines were relatively higher when the
pressure wheel force is set at 0.3, 6 or 12 N. The reason for this stems from the fact that when
welding of galvanized steel under a higher pressure wheel force (18 N), the gap near the
focal laser spot became too narrow to evacuate the high pressured zinc vapor; the zinc vapor
caused spatters that disturbed the stability of the plasma which affected the intensity of the
detected spectrum (Kong et al. 2012). The evolution of iron electron temperature within the
laser-induced plasma along the weld bead length is shown in Fig. 20. The iron electron
temperature was calculated by using the Boltzmann Plot method expressed by Equation (1)
(Kong et al., 2012, Griem, 1997 and Marotta, 1994):

Welding Processes 132


Figure 18. Schematic view of the setup for on-line monitoring the optical properties of plasmas during
laser welding


Figure 19. Spectrum of laser induced plasma captured by a spectrometer in the laser welding process
by various pressure wheel forces under a laser power of 4.0 kW and a welding speed of 50 mm/s
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 133

( ) ( )
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( )
2 1
1 1 2 2 1
ln
2 2 1 1 2
m m
e
m m m
m m m
E E
T
E I A g
k
E I A g

=
(
(
(

(1)
where Te is the plasma electron temperature, Em is the energy of the upper state, k is the
Boltzmann constant, Im is the emission line relative intensity, Am is the transition probability,
gm is the statistical weight, and m is the wavelength.
As shown in Fig. 20, under lower pressure wheel forces (0.3, 6, and 12 N), the electron
temperature showed lower intensity and less fluctuation compared to a higher pressure
wheel force (18 N). The presence of the zinc vapor induced spatters in the plasma which
increased the iron electron concentration which, in turn, increased the iron electron
temperature value (Kong et al., 2012).
Thus, there is a correlation between the optical emission of the plasma and zinc vapor
induced welding defects during the laser welding; and this optical signal could be further
used as feedback for the closed-loop control of the laser welding of galvanized steel with a
pressure wheel, which is shown in Fig. 21.





Figure 20. Electron temperatures of iron in laser-induced plasma captured by a spectrometer in the
laser welding process by various pressure wheel forces under a laser power of 4.0 kW and a welding
speed of 50 mm/s

Welding Processes 134

Figure 21. A schematic presentation of a closed-loop control system of the laser welding of galvanized
steel by controlling the clamping force
6. Conclusions
In this chapter, issues related to the laser welding of galvanized steels in a zero-gap lap joint
configuration are discussed. The authors recent research results on the laser welding of
galvanized steel in a lap joint configuration are reviewed. The different welding procedures,
namely, low power / low speed welding, two-pass laser welding and laser welding with a
pressure wheel are introduced for the laser welding of galvanized steels in a lap joint
configuration. It was found that acceptable weld quality could be achieved by a low power /
low welding speed procedure; however, the relatively low welding speed limits its
application in the industrial environment. A high quality weld could be obtained by
introducing a preheating pass with a defocused laser beam. By using the optical signal
acquired during the laser welding process as a feedback signal in the pressure wheel force
control, it is possible that a controllable clamping force could be a solution for achieving a
good weld quality without using a pre- and post- welding procedure.
Author details
Junjie Ma, Fanrong Kong and Radovan Kovacevic
*

Center for Laser-aided Manufacturing, Lyle School of Engineering, Southern Methodist University,
Dallas, TX, USA
Blair Carlson
General Motors R&D Center, Warren, MI, USA

*
Corresponding Author
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 135
Acknowledgement
This work was financially supported by the NSFs Grant No. IIP-1034652.
7. References
Akhter R., Steen W. M. and Cruciani D. (1988). Laser welding of zinc-coated steel.
Proceedings of the 5th International Conference on Lasers in Manufacturing, Stuttgart,
West Germany, September 13-14
Akhter R. and Steen W.M. (1990). The gap model for welding zinc-coated steel sheet. In:
Proceedings of the International Conference on Lasers Systems Application in Industry,
Torino, Italy, 219-236
AWS WZC (D19.0-72), Welding Zinc-Coated Steels (1972). American Welding Society.
Bakowski L., Beyer E., Herziger G. and Poprawe R. (1984). Development and optical
absorption properties of a laser induced plasma during CO2-laser processing. SPIE
Proceedings Series 455: 75-80.
Berlinger G. and Speranza J.J. (1987). U.S. Patent 4,684,779
Briand F., Chouf K. and Lefebvre P. (2008). Method and installation for laser welding with a
N2/He gas mixture, the N2/He content being controlled according to the laser power,
U.S. patent 7,385,158
Burns T. Weldability of a dual-phase sheet steel by the gas metal arc welding process.
Master Thesis, the University of Waterloo, 2009.
Chen W., Ackerson P. and Molian P. (2009). CO2 laser welding of galvanized steel sheets
using vent holes. Materials and Design (30): 245-251
Chung B.G., Rhee S. and Lee C.H. (1999). The effect of shielding gas types on CO2 laser
tailored blank weldability of low carbon automotive galvanized steel, Materials Science
and Engineering 272: 357-362
Dasgupta A., Mazumder J. and Bembenk M. (2000). Alloying based laser welding of
galvanized steel, Proceedings of International Conference on Applications of Lasers and
Electro Optics, Dearborn MI.
Dawes C. Laser welding: a practical guide, Abington publishing, 1992.
Dowden J. M. The theory of laser materials processing: Heat and mass transfer in modern
technology, Springer, 2nd edition, 2009.
Fabbro R., Coste F., Geobels D. and Kielwasser M. (2006). Study of CW Nd-Yag
laser welding of Zn-coated steel sheets. Journal of physics D: Applied physics
39:401-409
Forrest M. G. and Lu F. (2004). Advanced dual beam laser welding of zinc-coated steel
sheets in lap joint configuration with zero gap at the interface. 23th International
Congress on Applications of Lasers and Electro-Optics, ICALEO.

Welding Processes 136
Graham M. P., Hirak D. M., Kerr H. W. and Weckman D. C. (1996). Nd: YAG laser beam
welding of coated sheet steels using a modified lap joint geometry. Welding Journal
75(5): 162-170
Griem H. R. Principle of plasma spectroscopy. Cambridge Monographs on Plasma Physics.
Cambridge University Press, Cambridge, 1997.
Gu H. and Mueller R. (2001). Hybrid welding of galvanized steel sheet. 20th International
Congress on Applications of Lasers & Electro-optics, ICALEO, 130-139
Gu H. and Shulkin B. (2011). A practical use of humping effect in laser beam welding.
Journal of Laser Applications 23(1): 1-6
Gualini M.M.S., Iqbal S. and Grassi F. (2006). Modified dual-beam method for welding
galvanized steel sheets in lap configuration. Journal of Laser Applications 18(185): 185-
191
Heydon J., Nilsson K. and Magnusson C. (1989). Laser welding of zinc coated steel. In:
Proceedings of the 6th International Conference on Lasers in Manufacturing, 93-104
Homepage Fraunhofer - Gesellschaft: Http://www.fraunhofer.de/en.html
Imhoff R., Behler K., Gatzweiler W. and Beyer E. (1988). Laser beam welding in car body
making. Proceedings of the 5th International Conference on Lasers in Manufacturing,
Stuttgart, West Germany, September 13-14
Iqbal S., Gualini M.M.S. and Rehman A.U. (2010). Dual beam method for laser welding of
galvanized steel: Experimentation and prospects. Optics & Laser Technology 42(1): 93-
98
Kim C., Choi W., Kim J. and Rhee S. (2008). Relationship between the welding ability and
the process parameters for laser-TIG hybrid welding of galvanized steel sheets.
Materials Transactions 49(1): 179-186
Kennedy S.C. and Norris I.M. (1989). Nd-YAG laser welding of bare and galvanized
steels. SAE Technical Paper Series 890887, International Congress and Exposition,
Detroit, MI
Kong F., Ma J., Carlson B. and Kovacevic R. (2012). Real-time monitoring of laser welding of
galvanized high strength steel in lap joint configuration. Optics & Laser Technology
44(7): 2186-2196
Lacroix D., Jeandel G. and Boudot C. (1997). Spectroscopic characterization of laser induced
plasma created during welding with a pulsed Nd: YAG laser. Journal of Physics D:
Applied Physics, 81(10): 6599-6606
Li X., Lawson S. and Zhou Y. (2007). Novel technique for laser lap welding of zinc coated
sheet steels. Journal of Laser Applications 19(4): 259-264
Li X., Lawson S. and Zhou Y. (2008). Lap welding of steel articles having a corrosion
resisting metallic coating. U.S. Patent No. 2008/0035615 Al.
Loredo A., Martin B., Andrzejewski H. and Grevey D. (2002). Numerical support for laser
welding of zinc-coated sheets process development, Applied Surface Science 195(1-4):
297-303
Mitigating Zinc Vapor Induced Weld Defects in Laser Welding
of Galvanized High-Strength Steel by Using Different Supplementary Means 137
Ma J., Kong F. and Kovacevic R. (2012). Finite-element thermal analysis of laser welding of
galvanized high-strength steel in a zero-gap lap joint configuration and its experimental
verification. Materials and Design (36): 348-358
Marotta A. (1994). Determination of axial thermal plasma temperatures without Abel
inversion. Journal of Physics D: Applied Physics (27): 268-272
Mazumder J., Dasgupta A. and Bembenk M. (2002). Alloy based laser welding. U.S. patent
6,479,168.
Milberg J. and Trautmann A. (2009). Defect-free joining of zinc-coated steels by bifocal
hybrid laser welding. Production Engineering. Research and Development 3(1): 9-
15
Mitsubishi Co. (1993). U.S. Patent 5,618,452
Moeckel A., Stein H., Jucht H., (2003). Device for degassing laser welded seams Patent no
DE 10160156
Norris I., Hoult T., Peters C., Wileman P., (1992). Material processing with a 3 kW
Nd:YAG laser. Proceedings of Laser Advanced Materials Processing (LAMP),
Nagaoka, Japan
Pan Y. and Richardson I. M. (2011). Keyhole behavior during laser welding of zinc-coated
steel. Journal of physics D: Applied physics 44:045502
Park H.S. and Choi H.W. (2010). Digital Laser Welding System for Automobile Side Panel,
Laser Welding, Na, X.D., Stone (Ed.), ISBN: 978-953-307-129-9, InTech.
Ream S. (1991). Laser welding of zinc-coated steel. In: Laser applications in materials
processing and manufacturing. Society of Manufacturing Engineering (SME),
Southfield, MI
Ribic B., Palmer T.A. and DebRoy T. (2009). Problems and issues in laser-arc hybrid
welding. Int Mater Rev 54(4):223-44
Schmidt M., Otto A., Kgeler C. (2008). Analysis of YAG laser lap-welding of zinc coated
steel sheets. CIRP Annals - Manufacturing Technology, 57(1): 213-216.
Steen W. M. Laser material processing, 3rd edition, Springer, 2003.
Tzeng Y.F. (1999). Pulsed Nd:YAG laser seam welding of zinc-coated steel. Welding J 78(7):
238s-244s
Tzeng Y.F. (2006). Gap-free lap welding of zinc-coated steel using pulsed CO2 laser. The
International Journal of Advanced Manufacturing Technology, (29)3-4: 287-295
ULSAB-AVC Body Structure Materials, Technical Transfer Dispatch No. 6, May, 2001.
Wiese W.L. and Martin G.A. Wavelengths and transition probabilities for atoms and atomic
ions. Part II. Transition Probability. 1st edition, 1980
Xie J. and Denney P. (2001). Galvanized steel welding with lasers. Welding Journal 80(6): 59-
61
Yang S. and Kovacevic R. (2009). Welding of galvanized dual-phase 980 steel in a zero-gap
lap joint configuration. Welding Journal 88(8): 168-178

Welding Processes 138
Yang S., Carlson B. and Kovacevic R. (2011). Laser welding of high-strength galvanized
steels in a zero-gap lap joint configuration under different shielding conditions.
Welding Journal 90: 8s-18s
Section 2




Numerical Modeling of Welding Processes



Chapter 7




2012 Gao, licensee InTech. This is an open access chapter distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Numerical Modelling to Understand
Cracking Phenomena During Laser-GMA
Hybrid Welding Nickel-Base Superalloys
Zhiguo Gao
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/47395
1. Introduction
Laser welding is a highly efficient and precise welding method, which is greatly desired in
the automotive, aero and ship building industries. It elucidates some excellence, such as
higher depth to width ratio, concentrated heat input, minimal thermal distortion,
minimized weld fusion and heat affected zone, reduced post weld rework and possibly
joining widely dissimilar materials. But laser welding has also clearly demonstrated some
drawbacks. First of all, the cost of laser equipment and maintenance are high. Second, highly
reflective materials, such as aluminum, copper and gold, are more difficult to process with
some laser sources. Finally, the high welding speed leads to high solidification rate, which
in turn induces metallurgical problems, such as cracking, pores and brittle phase
structures.
Laser-gas metal arc (GMA) hybrid welding is one of the state-of-the-art technologies and is
designed to overcome problems commonly encountered during either laser or arc welding.
Hybrid welding offers some additional advantages over mere laser and arc welding. Firstly,
the combination of a laser beam and gas metal arc in welding will reduce the propensity for
cracking, brittle phase formation and porosity. Secondly, the feeding wire modifies the
metallurgy of the weld. Thirdly, the effect of the coupled process exceeds the effects of the
laser and arc heat sources taken separately, and the laser beam located in front of the GMA
weld pool could suppress hump formation during high travel speed (Claus et al., 2005; Gao
et al., 2009; Ribic et al., 2009).
Nickel-base superalloys are extensively used as aeronautical material due to their high
temperature properties. However, commonly used types of commercial welding for nickel-

Welding Processes 142
base superalloys have three major hurdles that need to overcome in order to make
refurbishment and repair feasible. First, the single crystal nature of nickel-base superalloys
is easily lost during welding due to stray grain formation. This phenomenon is called a
columnar-to-equiaxed transition (CET). Secondly, polycrystalline nickel-base superalloys
are very prone to cracking during welding, and these cracks are mostly thermally induced
cracks which originate in the grain boundary in the heat affected zone (HAZ). The
combination of welding stress and increased concentration of gamma-prime are thought to
be the cause of hot cracks in the HAZ. Finally, on-equilibrium solidification, elemental
partitioning and subsequent solid state transformation can yield non-ideal microstructures
and will not produce material with the desired properties.
It has been found that the mechanical strengths at elevated temperatures of columnar
grains are better than those of gas turbine blades with equiaxed grains. Columnar grains
are desirable, and equiaxed grains are treated as casting defects in directionally
solidified or single crystal castings. Some researchers have implemented theoretical
analyses on the solidification behaviours of nickel-base superalloys. Hunt (1984) and
Gumann et al. (2001) theoretically analyzed stray grain formation to consider
the parameter for describing the extent of nucleation and growth for stray grain
formation ahead of the advancing solidification front. represents the volume fraction
of equiaxed grains and varies from 0 to 1. The value of can be calculated by the
following equation:

( )
4
1
0
3
1 3ln 1
n
n
N
G
a
V n


=
`
+ u

)
t
(1)
where G is the thermal gradient, V is the growth velocity, a and n are material constants and
N0 is the nucleation density. a=1.2510
6
K
3.4
m
-1
s
-1
, n=3.4 and N0=210
15
m
-3
are used for a
similar nickel-base superalloy. When constitutional supercooling is minimal, =0, the weld
microstructure is free of stray grains and maintains its single crystal nature. When =1,
solidification is equiaxed. A critical value of that corresponds to a stray grain free
microstructure is hypothesized to be 0.0066.
Gumann et al. (2001) derived a criterion based on the G
n
/V ratio which states that the
microstructure is columnar when the following condition is satisfied everywhere in the
weld melt pool.

( )
3.4
24 3.4 4.4
2.7 10 K /m s
G
V
> (2)
Park et al. (2003) used the degree of constitutional supercooling (CS) at the solidification
front as a metric to assess the effects. The basic criterion to avoid CS, which leads to the
growth of new grains at the solidification front, is given by:
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 143

cr
G G
V V



(3)
where (G/V)cr is a critical value of G/V, which is proportional to T/DL (T is the
solidification temperature range, and DL is the diffusion coefficient of solute in the liquid).
In cases where the planar solid/liquid interface breaks down into cellular or dendritic
solidification, the theory of CS predicts fairly closely. However, no theory can predict the
(G/V)cr for the CET.
Liu and DuPont (2004, 2005) studied the effects of melt-pool geometrical parameters on
crystal growth and microstructure development during laser surface melting of single-
crystal alloys.
Thanks to the development of welding techniques, much progress has been made to discuss
cracking phenomena in the laser join nickel-base superalloys (Egbewande et al., 2010; Hong
et al., 2008; Liu et al., 2011). Recently, feasible finite element models were established to
provide substantial insights into cracking and how thermal gradient, transient stress and
strain were developed in the laser welding process (Dai&Shaw,2001; Hee et al.,2010;
Hu&Richardson,2006;Long et al.,2008;Luo et al.,2002; Nishimoto et al., 2002; Wang et al.,
2004; Yann et al.,2010 ;Yilbas et al.,2010).
As mentioned previously, albeit laser welding is achieved by rapid heating and rapid
cooling, solidification cracking and liquation cracking are prone to be generated in the weld
and HAZ. The effect of weld pool shape on the solidification cracking and HAZ cracking
susceptibility still remain obscure in many respects for the laser welding. Some
experimental conclusions from previous literature have proved that the number of micro
cracks at the fusion line in the weld cross-section is reduced with increasing heat input for
laser-arc hybrid welding, and cracking might be prevented by adjustment of filler material
and heat input for the laser-arc hybrid welding. However, limited information and rationale
are available in the literature about how the weld solidification cracking and HAZ crack
susceptibility are minimized during laser-arc hybrid welding.
In this chapter, in order to properly explore the capability of the new welding process, i.e.
laser-arc hybrid welding, it is necessary to know and understand how laser-arc hybrid
welding affects solidification cracking and liquation cracking. A major factor control
susceptibility to solidification cracking and liquation cracking are stray grain formation and
stress-strain generation during weld pool solidification, respectively. It is important to study
and understand stray grain formation and stress-strain generation mechanism during laser-
arc hybrid welding, whereas these information are currently scarce in the literature.
2. Solidification cracking and liquation cracking modeling
In this section two kinds of theoretical analysis models are developed to understand
cracking phenomena, one is modelling analysis of hybrid laser-arc welding of single-

Welding Processes 144
crystal nickel-base superalloys and the other is modelling analysis HAZ liquation cracking
of laser hybrid welding polycrystalline nickel-base alloy and schematic diagram is shown
in Fig.1.


Figure 1. Schematic diagram of cracking development and theoretical analysis for the laser-GMA
hybrid welding nickel-base superalloys
2.1. Solidification cracking modeling for single-crystal superalloys welding
2.1.1. Mathematical model description
To execute the analysis on the dynamic behavior and solidification of molten pool, three-
dimensional essential governing equations, such as continuity, Navier-Stokes, energy and
VOF method equations are iteratively solved with the assumption that the liquid flow is
Newtonian, laminar and incompressible.
Continuity equation
V
l s
m (4)
Momentum equation
Weld pool geometry









Cracking development with metallurgical and mechanical
factors
Welding conditions
Temperature distribution
Fluid flow pattern
Microstructure evolution
Crystallographic growth
Solid-state transformation
Dendrite grain structure
Stress-strain development
Thermal stress/strain
Elastic stress/strain
Plastic stress/strain
Weld solidification cracking HAZ liquation cracking
Thermal elasto-plastic model
Strain rate
Strain rate gradient
Microstructure development model
Dendrite growth velocity
Thermal gradient
Columnar-to-equiaxed transition
Stray grain formation
Strain distribution
Arc parameters:
Current, voltage, metal transfer
Laser parameters:
Laser power, welding speed
The distance between arc and laser beam
Theoretical analysis Theoretical analysis
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 145

2
V
V V V V F V
| | c
+ V = V V +
|
c
\ .

l s
P+m K
t
(5)
Energy equation

( )
( ) ( )
l l s
k T U
t
| |
c
+ V = V V + |
|
c
\ .

U
U V
(6)
( ) ( )
1
s st
T
U c T dT f h = +
}
(7)
where V is the molten metal velocity,
s
m is a mass source term, P is the hydrodynamic
pressure, is the dynamic viscosity, F is the body force, l is the fluid density, U is internal
energy per unit mass, kl is thermal conductivity, T is a local temperature,
s
U

is an energy
source term due to mass source term, K is the drag coefficient for the porous media model, fs
is the solid fraction, and c (T) is specific heat.
According to the kinematic motion equation, the VOF function moves according to the
velocity field in the fluid, as shown by Equation (8).
( )
s
F
F F
t
c
+ V =
c

V
(8)
For a single fluid, incompressible problem F represents the volume fraction occupied by the
fluid. Thus, fluid exists where F=1, and void regions correspond to locations where F=0.
s
F

represents the change of volume fraction of fluid associated with the mass source .
The local temperature gradient Ghkl and dendrite growth velocity Vhkl can be calculated to
predict the microstructure development across the entire solidification interface. The
solidification front produced at the rear part of the weld pool is identified by a solidus
curve. The effect of keyhole penetration on weld pool geometry is considered in the analysis
of the microstructure development. Fig. 2 illustrates a schematic diagram of the weld pool
and associated geometry variables, which illustrate the relationships among the welding
velocity
b
V

, the solidification front unit vector n

, and the dendrite growth velocity


hkl
V

. is
the angle between the solidification front normal n

and a particular dendrite growth


direction [hkl]. is the angle between the surface normal n

and the welding direction, and


is the angle between the y-axis and the projection of n

on the y-z plane. Lal is the distance


between the laser beam and GMA, cl and ca are the penetration of laser and arc welds,
respectively.
Based on previous analytical methods(Liu&DuPont,2004,2005; Rappaz et al.,1990), a
mathematical model is extended for the calculation of stray grain fraction during
solidification microstructure development in the hybrid weld pool. The 3-D melt-pool
geometry on the rear part of the weld can be described in an ellipsoid formula.

Welding Processes 146

2 2 2
2 2 2
1
y x z
a b c
+ + = (9)
where a, b and c are the half-axes of the ellipsoid in the x, y and z directions, respectively.
As well, a is the length between the location of the maximum width and trailing point, b is
the maximum width of the arc pool, c is the maximum depth and in consideration of the
solidification history of weld pool penetration, cl is used in this model.
The relationship of the velocity of the dendrite tip,
hkl
V

, along a specific [hkl] direction is


given by:

Figure 2. Schematic diagram of weld pool geometry mathematical model and associated variables
(Gao&Ojo,2012a)

cos
cos cos
n
hkl b
V
V V = =


u

(10)
where
n
V

is the growth rate of the normal to the solidification front at the liquid-solid
interface.
The unit component based on the weld pool normal in the x, y and z axes are determined by
( )
cos , sin cos , sin sin n =

u u u
(11)
In order to determine the growth velocity along the [hkl] dendrite growth direction (one of
the six <100> directions), the value of cos can be determined from the relation:
cos cos (sin )(cos ) (sin )(sin ) h k l = + u u u (12)
The thermal gradient along the dendrite growth direction,
hkl
G

, is given by the expression:



cos
hkl n
G G =

(13)

n

O
Y
X
c
l

a

Z

L
al

c
a

b
n

b V

hkl V

Numerical Modelling to Understand Cracking Phenomena


During Laser-GMA Hybrid Welding Nickel-Base Superalloys 147
where
n
G

is the total thermal gradient direction in the x, y and z directions.


The expression is solved for the volume fraction of stray grains as a function of Ghkl and
Vhkl which yields:
1
S
e u = (14)

( )
( )
3
3 3.4
19 0
1 3.4
4 1
2.356 10
3
1 /
hkl
n
n
hkl
N V
S
G
n G aV
| |
| |
|
| = =
|
|
| \ . +
\ .
t
(15)
The area-weighted average across the entire solidification interface, u , is given by the
expression:

k
k
k
k
A
A
u
u =

(16)
where the variable k corresponds to the cross-section along the x axis, and Ak and u are the
areas and average values for each cross-section, respectively. More information about this
numerical model can be reviewed by the reference (Gao&Ojo,2012a).
Thermal properties of typical nickel-base alloy used in this model are directly taken from
values produced in the previous literatures(Anderson et al.,2010; Banarjee&Overfelt,1999;
Bonifaz&Richards,2009; Brooks et al.,1996; Luo et al.,2002; Pottlacher et al.,2002; Zeng,2006),
and are presented in Table 1.

Material property Value Unites
Density of liquid 7578.9 Kg m
-3

Density of solid 7742.5 Kg m
-3

Viscosity of liquid 0.0074 Kg.s/m
Solidus temperature 1507 K
Liquidus temperature 1613 K
Specific heat of solid 430 J kg
-1
K
-1

Specific heat of liquid 700 J kg
-1
K
-1

Thermal conductivity of solid 11.4 Wm
-1
K
-1

Thermal conductivity of liquid 28.7 Wm
-1
K
-1

Coefficient of thermal expansion 1.3310
-5
K
-1

Latent heat of fusion 1.4510
5
J kg
-1

Latent heat of vaporization 6.410
6
J kg
-1

Surface tension 1.8 N m
-1

Surface tension gradient -1.3710
-3
N m
-1
k
-1

Table 1. Nickel-base alloy properties used in this model

Welding Processes 148
2.1.2. Results and discussion
The material adopted for this simulation is a nickel-base superalloy with a thickness of 3
mm. The focal plane of the laser beam is placed on the top surface of the base metal. Arc
droplets are assumed to be steadily generated through a certain wire feed rate at an initial
speed of 1.5m/s, and initial temperature 1613K. The size of the droplet mainly depends on
the diameter of the feeding wire, and the generation frequency is determined by the wire
feed rate and wire diameter. The distance between the arc source and laser beam is 3 mm,
and the different welding conditions are listed in Table 2.


Current
(A)
Voltage
(V)
Laser
power
(W)
Welding
velocity
(m s
-1
)
Droplet
frequency
(Hz)
Arc heat
distribution
parameter(mm)
Weld1 120 21 3667 0.022 217 2.23
Weld2 150 21 3667 0.022 269 2.33
Weld3 180 21 3667 0.022 304 2.45
Weld4 120 21 3667 0.028 217 2.23
Weld5 120 21 3667 0.034 217 2.23
Weld6 120 21 4000 0.022 217 2.23
Weld7 120 21 4332 0.022 217 2.23
Table 2. List of Hybrid welding parameters used for Nickel-base superalloy (Gao&Ojo,2012a)
2.1.2.1. Role of welding parameters on microstructure development
The properties of weld metal are affected by solidification behavior parameters, such as
growth rates, temperature gradients, undercooling and alloy constitution, which are often
useful to determine the development of microstructures in weld zones (Farzadi et al., 2010;
He et al., 2005; Roy et al., 2006). In this study, the effect of undercooling has been simplified
and thereby the solidification parameters, such as cooling rate, thermal gradient, and
solidification velocity, are computed by the weld pool geometry considering only the heat
transfer and fluid flow.
Since the shape of the weld pool remains constant under a steady-state condition, the
solidification rate can be calculated with position along the fusion boundary. The steady state
solidification rate, Vn, is related to the welding velocity in the following formula relation:
cos
n d
V V (17)
where is the angle between the normal to the solid/liquid boundary and the welding
direction, and Vd is the welding velocity. The solidification rate increases from the edge of
the weld pool and pool bottom, where is nearly 90, to the weld centerline, where the
velocity equates Vd along the fusion boundary (Farzadi et al., 2010).
To remove the constitutional undercooling ahead of the growth front for columnar grain
growth, a high thermal gradient is required and low solidification velocities are desired
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 149
(Bussac & Gandin, 1997; Dong, 2007). The values of the temperature gradient and
solidification rate vary with position along the entire solidification interface of the weld
pool. The density of temperature isotherm in the maximum width is greater than that of the
trailing portion of the pool boundary, so the temperature gradient is higher at the fusion line
than the centerline. The ramifications of cooling rates (GR) along the solid/liquid interface
with different welding conditions are presented in Fig.3.
The cooling rate is higher at the maximum width of the fusion boundary and lower at the
weld centerline for each case in Fig.3 (A). It is observed that the average range of the cooling
rate is 410
3
-2.410
4
K/s. The magnitude of the cooling rate decreases with an increase in the
arc current and laser power, while an opposite effect will occur with a decrease in welding
speed. In general, increasing the welding speed or decreasing welding heat sources power
will increase the cooling rate, and the latter will result in an increase of residual stress and
strain. In addition, the solidification velocities will be high and this may lead to more
extensive non-equilibrium solidification and partitioning at high welding speeds (Rai et al.,
2007; Vitek, 2005).
The cooling rate along the solid/liquid boundary of penetration in the conjoint zone is
shown in Fig.3 (B). It is observed that the magnitude of the cooling rate is 10
5
-710
5
K/s,
which is approximate to the typical laser welding range of 10
4
-10
6
K/s (Ral et al., 2009). It is
higher at the bottom of the keyhole than that at the weld crown solid/liquid boundary. This
distribution is consistent with higher convective heat transfer near the pool surface than the
weld root. The Marangoni convection brings fluid from the weld center to the peripheral
solid/liquid interface, which yields a lower thermal gradient on the weld pool surface. It can
be further found that the value of the cooling rate near the bottom of the keyhole
dramatically increases with increase in the arc current, welding speed and laser power, and
the effect of laser power is the most obvious. Meanwhile, the magnitude of the cooling rate
near the maximum depth of the weld pool is much higher than that of the rear part of the
weld pool.

Figure 3. The calculated cooling rate along the solidification interface for different welding conditions,
(A) Weld pool surface, (B) Weld penetration direction (Gao&Ojo, 2012a)

Welding Processes 150
Secondary dendrite arm spacing (SDAS) is estimated from the cooling rates (GR).The
dendrite arm spacing increases from the interface of maximum weld width to the weld
centerline, and decreases from the weld crown to weld root, as shown in Fig.4. These
explicate that the finer arm spacing is always found near the fusion boundary on the weld
crown or part of the weld root, and the coarsest structure occurs in the center of the weld
crown.
As the arc current and laser power were increased, the dendrite trunk spacing along the rear
of the solid/liquid interface of the weld pool increases and the magnitude of the SDAS is
between 2.88 and 4.2 m. Dendrite arm spacing decreases with increasing velocity, more
obvious for higher welding speed, and the characteristic magnitude of SDAS is between 2.18
and 3.2 m. The calculated dendrite arm spacing along the solidification interface has the
same tendency as the measured values, but the former is slightly greater than the latter,
especially near the fusion boundary, see Fig.4 (A).
The SDAS near the solid/liquid boundary along the weld penetration is presented in Fig.4
(B). The dendrite trunk spacing near the root part of the weld gradually decreases with
increase in the arc current, welding speed and laser power. The calculated dendrite arm
spacing along the direction of weld penetration agrees reasonably well with the measured
value near the lower region of the keyhole, and the experiment result is higher in value
than that calculated on the weld crown. Hybrid welding allows for the weld metal to
remain at a high temperature for a longer time and cool at lower rates compared to
approximated laser power performed at the same welding speed (Moore et al., 2004). This
results in a coarser dendritic microstructure region where the laser and arc beams
simultaneously interact, which is at the weld crown center rather than the solidification
boundary.


Figure 4. Predicted and experimental results of SDAS along the weld pool solidification interface, (A)
Weld pool surface, (B) Weld penetration direction
(A)
(B)
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 151
Since the effect of the crystallographic orientation of the weld on the overall tendency to
form stray grains is minimal, the weld direction [100] and sample surface plane (001) are
specified for the sample orientation. The angles (, ), which characterize the normal to the
solidification front, are determined from three-dimension weld geometry, and dendrite
growth pattern transitions can occur along the solidification interface. In consideration of
the average growth velocity along the solidification interface front at each analysis cross-
section, the dendrite orientation velocities of [100], [010] and [001] are presented in Fig.5.
The , , terms are used to calculate the dendrite growth velocity by Eqs. (10) and (12).
The velocities of the [100] dendrites are always equal to those by welding, regardless of the
weld pool shape for the (001)[100] orientations (Rappaz et al.,1990), and average dendrite
tip velocity in [010] and [001] is higher than that of the [100] orientation along the rear part
of the solid/liquid interface. The average dendrite tip velocity of the [001] orientation is
much larger than that of [010], and the latter is higher than that of [100] near the
solidification interface. It can be underlined that > and the dendritic orientation does not
coincide with the solidification front for [010] and [001]. The discrepancy between and is
larger in [001] than [010].At the weld center, the dendrite growth velocities of [010] and [001]
are lower than those at the solid/liquid boundary. For case (A), there are obvious transition
positions among the solidification interface with differences in the average dendrite velocity
for the [001] orientation. The average dendrite growth velocity of the [010] and [001]
orientations of weld2 are lower than the other cases near the solidification interface. For case
(B), the solidification time is much less at high welding speeds, and dendritic arrays do not
have sufficient time to establish steady state spacing. The average dendrite tip velocity in
the [010] and [001] orientations are irregularly volatile along the solidification front at the
highest welding speed. It is also observed that the welding speed affects the magnitude of
the dendrite growth velocity for the [001] orientation near the solidification interface. For
case(C), the average dendrite tip velocity for the [001] orientation monotonically increases
with increase in laser power, and its magnitude is the largest near the fusion line with the
highest laser power.
2.1.2.2. Role of welding parameters on stray grain formation
The area-weighted average values of stray grain formation tendency with different welding
conditions is shown in Fig.6, and partial columnar growth with the value of is below 0.5
and above 0.0066 with increase in arc current or welding speed. These welding parameters
are beneficial for avoiding CET during the entire welding process. The effect of welding
speed on is more complex. The latter initially increases with increasing welding velocity
before decreasing at higher speeds. This can be explained through the changes in
temperature gradient transition with different welding speeds. As the welding speed is
increased, the temperature gradient is generally affected by two opposing factors (Grong,
1997). Thus, the magnitude of will subsequently decrease with a further increase in the
welding speed. The results from Anderson et al. (2010) and Vitek (2005) are similar to those
illustrated here, in which the fraction of stray grain formation is generally observed to reach
the maximum at an intermediate speed, and then always decreases as the speed increases.

Welding Processes 152
Hitherto, literature mainly focused on wide and shallow pool shapes and higher laser power
conditions were omitted due to limitations in simulation and method (Anderson et al., 2010;
Gumann et al., 2001;Vitek,2005).From the investigations of this model, the effect of higher
laser power is slightly negative on SG formation. Laser power is the primary factor in weld
penetration and weld pool geometry which determines the shape of the liquid/solid
interface and the slope of the weld pool edges. The local angle along the solidification
interface in the nail-head shaped weld contributes more to stray grain formation. Hence, the
typical keyhole geometry in a hybrid pool is more prone to create CET in the weld pool





Figure 5. Dendrite tip velocity distribution around rear part of weld pool with different welding
conditions, (A) variation of arc current, (B) variation of welding velocity, (C) variation of laser power
(Gao&Ojo,2012a)
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 153
1000 1200 1400 1600 1800 2000 2200 2400 2600
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
Equiaxed grain
Columnar grain

A
r
e
a
-
w
e
i
g
h
t
e
d

a
v
e
r
a
g
e

s
t
r
a
y

g
r
a
i
n
s
Net heat input(J/cm)
Tendancy with laser power Caculated value
Tendancy with welding speed Caculated value
Tendancy with arc current Caculated value

Figure 6. Area-weight average values of stray grain formation distribution as a function of heat
input(Gao&Ojo,2012a)
when considering the distribution of in the three directions from the above simulation,
whereas the net effect of higher laser power on CET can be verified by the experimental
work of Anderson et al. (2001), who used similar welding power intensity with an electron
beam.
From the different tendencies of summarized by the calculated value, higher welding
speed and lower arc current or laser power ameliorate to minimize/avoid stray grain
formation. In contrast, slower welding speed and higher arc current or laser power
exacerbate stray grain formation, and abundant equiaxed dendrite will be yielded. Based on
the above analysis, welding conditions should be optimized to minimize the overall
likelihood of forming stray grains.
2.2. Liquation cracking modeling for polycrystalline superalloys welding
2.2.1. Mathematical model description
In consideration of tractable object symmetry, a 1/2 model including the weld metal,
deposition filler material and base metal is adopted. A sequential thermal-mechanical finite
element analysis is performed in present study based on ANSYS 13 code. Specifically, 2D
thermal solid plane55 is used to simulate the temperature field and 2D structure element
plane42 is then used to calculate the stress-strain field with the nodal temperature extracted
from the thermal analysis as the load. All the material properties used in the model are
displayed in Fig.7.
In this model, the laser-GMA hybrid process employs heating the base metal in a localized
zone with sequent transient heat sources. The effect of the synergistic interaction between
laser and GMA on the heat flux distribution is not considered in this simulation. The
distribution of laser beam energy is assumed to be Rotary-Gauss body heat source, which is

Welding Processes 154
suitable for nail-head shape with large depth to width ratio of molten pool along the
workpiece thickness direction (Wu et al., 2004).The distribution of the GMA energy is
assumed to be a Gaussian flux over the workpiece surface.

Figure 7. The physical and mechanical properties used for the simulation (Gao&Ojo, 2012b)
The crystallographic orientation of the dendrites is not taken into account and they are
assumed to grow in a direction that is always perpendicular to the growth front. Local
growth velocity of dendrite along the solidification front as a function of undercooling is
represented by the following function (Hunziker et al., 2000):

{ }
n
kin
v T a T A = A (18)
where n=3.05, a=1.275410
-7
ms
-1
K
-3.05
and the undercooled liquid is the region between the
growth front and liquidus temperature. Using growth rate R and the temperature gradient
G along the fusion boundary, the cooling rate (GR) can be calculated.
For the case of nonlinear material, the definition of total strain is given:

{ }
{ } { } { }
pl el th
= + + c c c c (19)
where
el
is the elastic strain vector,
th
is the thermal strain vector,
pl
is the plastic strain
vector but the creep strain vector and swelling strain vector are ignored.
As mentioned above, the thermal elasto-plastic material model, based on von Mises yield
criterion, temperature dependent mechanical properties and linear kinematic hardening
rule, is considered. The incremental forms of stress-strain relation are described as (Akbari
& Sattari-Far, 2009; Chang & Lee, 2009; Kong & Kovacevic, 2010):

{ }
ep th
d D d D dT
( (
( =


o c (20)

ep p e
D D D
( ( (
= +

(21)
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 155
where d is the stress increment, d is the strain increment, [D
e
] is the elastic stiffness
matrix, [D
p
] is the plastic stiffness matrix, [D
th
] is the thermal stiffness matrix and dT is the
temperature increment. More information about this numerical model can be reviewed by
referring to Gao ( 2012).


Laser
power (kW)
Arc power
(kW)
Welding
speed
(m/min)
Time of laser
heating (s)
Time of arc
heating (s)
Time interval
for hybrid (s)
Laser 2 - 3 0.012 - -
Laser 4 - 3 0.012 - -
Laser 6 - 3 0.012 - -
Laser 4 - 2 0.018 - -
Laser 4 - 4 0.009 - -
Laser 4 - 6 0.006 - -
Laser-GMA 2 4.75 3 0.012 0.05 0.02
Laser-GMA 4 4.75 3 0.012 0.05 0.02
Laser-GMA 6 4.75 3 0.012 0.05 0.02
Laser-GMA 4 4.75 2 0.018 0.075 0.03
Laser-GMA 4 4.75 4 0.009 0.0375 0.015
Laser-GMA 4 4.75 6 0.006 0.025 0.01
Table 3. Design laser and laser-GMA hybrid welding conditions (Gao&Ojo, 2012b)
Predictions for the cracking susceptibility have been estimated for a wide range of welding
conditions. Two types of welding processing, three kinds of laser power and four kinds of
welding speed have been investigated, which result in a total twelve laser fabrication
conditions, as shown in Table 3.There is a time interval for laser-arc hybrid welding, which
sequentially heats the base material due to the distance between laser beam and arc source.
2.2.2. Results and discussion
2.2.2.1. Weld pool solidification characteristics
The HAZ cracking in the waisted zone over the bead cross-section is attributed to the
greater amount of grain boundary liquation during the cooling process. The solidification
completion temperature of liquated grain boundary is around 1383K, which affects the
Laves cluster dissolution and grain boundary liquid phase (Nishimoto et al., 2002). The
shape of weld pool and cooling rate along localization fusion boundary derived from
temperature isothermal are shown in Fig.8. The abscissa means the distance from the surface
to the root of the weld bead along the fusion boundary. The left ordinate value indicates the
weld geometry, which is the distance from the weld bead centerline and the right ordinate
value is the cooling rate.

Welding Processes 156

Figure 8. Correlation between the shape of weld pool and cooling rate with welding parameters for
laser welding , (A) variation of welding speed, (B) variation of laser power
The results indicate that the cooling rate is a function of position along the fusion boundary
among the weld fusion zone and varies simultaneously with the penetration shape. The
shape decay of temperature in the surface region is attributed to the heat loss from the
surface region due to heat transfer of conduction and convection in this region. Temperature
decay in the initially high temperature region at the root of the keyhole during the laser
beam moving away, which is due to the attainment of high temperature gradient in this
region, results in faster solidification. The neck zone over the bead cross-section is more
liable to heat stagnation than in other regions and undergoes lower cooling rates than other
parts. That means that the maximum solidification time exists at the neck zone of the weld
shape and the minimum solidification time occurs at the root of the keyhole. Thus, there is
obvious solidification time transition region along the fusion boundary line. There is a steep
cooling rate in the vicinity of the neck zone. With an increasing welding speed, the cooling
rate in the neck zone gets steeper and weld pool geometry substantially decreases,
respectively. Correspondingly, the shapes of weld pool synchronously enlarge with a laser
power increase and the concomitant cooling rates considerably decrease.
Correlation between weld pool geometry and cooling rate of the hybrid pool characteristic
are shown in Fig.9. The difference in these relationships between laser and hybrid laser arc
welding are compared, and several conclusions can be drawn from the proposed study.
First, with incorporating the effect of filler material on the weld bead cross-section, the weld
pool geometry variation with welding speeds of laser hybrid welding are different from that
of laser welding. Employing an additional heat and deposition metal lead to an extended
region of weld width and increased neck zone radius of curvature. The volume of weld filler
metal strongly depends on the welding speed and decreases with an increasing welding
speed. Second, the rational link between cooling rate and weld pool geometry suggest that
the neck zone over bead cross-section is more liable to heat stagnation and has larger
solidification time, therefore, the minimized cooling rate occurs more readily at locations
where it is underside of neck zone, which is especially observed in the neck-like waisted
(A)
(B)
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 157
zone of lower heat input of laser-GMA hybrid welding as well as laser welding. Third, laser-
GMA hybrid welding alters the temperature distribution of the weld fusion zone and
substantially enhances the cooling rate near the weld bead neck zone. The magnitude of
cooling rate is lower than that of laser welding. Cooling rate attains low value in the region
of neck zone due to higher solidification time, and obviously increases near the bottom of
the keyhole. The relationship between the weld pool shape and the cooling rate elucidates
clearly localized solidification behavior and enables the cause of liquation cracking to be
explained.

Figure 9. Correlation between the shape of weld pool and cooling rate with various welding conditions
for laser-arc hybrid welding, (A) variation of welding speed, (B) variation of laser power
2.2.2.2. HAZ cracking analysis
Constitutional liquation theory and grain boundary segregation mechanism are associated
with weld HAZ liquation cracking. Nishimoto et al (2002) explained the liquation cracking
mechanism in laser welds of Inconel 718. The molten liquid of the Laves phase eutectically is
liquated at a temperature below the solidus temperature of the matrix and infiltrates along
the grain boundaries. Tensile plastic strain/stress induced by thermal shrinkage is imposed
on the liquid film during the subsequent cooling process and cracking will occur with an
attendant buildup of higher strain/stress across the grain boundaries.
Despite the fact that the HAZ cracking metallurgical mechanism of nickel-base alloy is
revealed, the limited information is available in the literature about the explanation of the
relationship between welding conditions and cracking susceptibility during the laser and
hybrid laser-GMA welding.
For the purpose of comparison of crack susceptibility with various welding conditions and
different welding types, von Mises stress and 1
st
principal strain are not sensitive to reflect
these discrepancies (Gao, 2012). Quantitative evaluation of solidification crack with behavior
of ductility near the solidification boundary can be calculated by crack susceptibility strain
rate, which is represented:
(A)
(B)

Welding Processes 158

1
s
t


(22)
where 1 denotes total 1
st
principal strain along the fusion boundary, ts denotes the lapse
time of solidification boundary from liquidus temperature to crack sensitive temperature,
1383K.

Figure 10. Correlation between the shape of weld pool and strain rate with various welding conditions
for laser welding, (A) variation of welding speed, (B) variation of laser power(Gao&Ojo, 2012b)
Relations among crack susceptibility strain rate and pool shape of laser weld are shown in
Fig.10. As a whole, it is worth noting that the value of the geometry of weld pool is nearly in
the same order of the crack susceptibility strain rate, and the curve shapes of crack
susceptibility strain rate conspicuously change with various welding conditions. It should
be noticed that there is a concave shape, which suddenly decreases and then increases in the
strain rate curve adjacent to the neck region of weld pool shape that means it is more
susceptible to initiate crack in the stage of solidification. After the neck zone, strain rate
increases until it is at the root of the keyhole and reaches a maximum value. The reason for
this gradual increase in strain rate is thought to be due to the interaction between strain
distribution and solidification characteristic within the weld pool. It can be estimated that
greater welding speed and smaller laser power, namely less heat input results in the steeper
curve of crack susceptibility strain rate, which is prone to engender and propagate cracks in
the neck zone. Thus, the crack susceptibility strain rate itself is a major parameter
determining the tendency of crack susceptibility.
As mentioned before, the effects of the welding conditions on the solidification crack
susceptibility are interrelated. In contrast, laserGMA hybrid welding can minimize the risk
of variation of strain rate in the neck zone of the weld bead, and the value of these ductility
curves are obviously smaller than that of laser welding, as shown in Fig.11. The strain rate
becomes sharp near the neck region with an increasing welding speed or decreasing laser
power.
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 159

Figure 11. Correlation between the shape of weld pool and strain rate with various welding conditions
for laser-arc hybrid welding, (A) variation of welding speed, (B) variation of laser power(Gao&Ojo,
2012b)

Figure 12. The strain rate gradient near the neck region with different welding conditions, (A) variation
of welding speed, (B) variation of laser power
Comparison of strain rate gradient near the neck region under different welding conditions
is illustrated in Fig.12. It is worth noting that the strain rate gradient decreases with heat
input increase. The values of strain rate gradient of laser welding are on the whole higher
than those of hybrid laser arc welding. It clearly reveals that laser welding is more
susceptible to liquation cracking and the crack is located more in the neck region during low
heat input. Therefore, the solidification cracking relating the liquid film stage in cracking
sensitive temperature range can be explained from the view point of the contribution of
strain rate gradient. From the above results, using strain rate gradient for evaluation of the
susceptibility to HAZ liquation cracking provides theoretically understanding of how laser-
arc hybrid welding produces less susceptibility to liquation cracking compared to laser
welding as previous reported experiment results (Stelling et al., 2005).

Welding Processes 160
3. Conclusions
These models provide valuable insight into the metallurgical and mechanical driving force
for cracking. These results will be used to predict the effect of welding conditions on the
potential for weld solidification cracking and HAZ liquation cracking and to interpret the
experimental results during laser-GMA hybrid welding single-crystal and polycrystalline
nickel-base superalloys. It should be noted that higher welding speed and lower laser power
or arc current are beneficial for minimizing stray grain formation for laser-GMA welding
single-crystal nickel-base superalloys , but generally lower heat input exacerbate HAZ
liquation cracking for laser-GMA welding polycrystalline nickel-base superalloys. These
promising results will also be used to identify preferred welding conditions that minimize
stray grain formation, local stress-strain and associated cracking. For the future research,
many efforts will focus on the different welding parameters on the cracking trend with
optimized tactics and substantiate them by experiments.
For the solidification cracking modeling of welding of single-crystal nickel-base superalloys,
there are some conclusions:
1. In almost all cases, the cooling rate is higher at the maximum width of the fusion
boundary and at the bottom of the keyhole, while lower at the weld centerline and weld
crown solid/liquid boundary. Meanwhile, it is shown that the finer arm spacing is
always found near the solid/liquid boundary of the weld crown or the weld root, and
the coarsest structure occurs in the center of the weld crown;
2. The average dendrite tip velocity of the [001] orientation is much greater than that of
[010], and the latter is higher than that of [100] near the solidification interface for the
(001)[100] orientations. At the weld center, the average dendrite growth velocity of
[010] and [001] is lower than that at the solid/liquid boundary;
3. Higher welding speed and lower laser power or arc current are beneficial for
minimizing stray grain formation, when taking into consideration, on average, the
tendency for stray grain formation over the entire weld pool;
For the HAZ liquation cracking modeling of welding of polycrystalline nickel-base
superalloys, there are also some conclusions:
4. The neck zone radius of curvature in the laser hybrid pool is larger than that of the laser
pool, which means the neck region is more liable to heat stagnation and has a smaller
cooling rate. The minimized cooling rate occurs more readily at the underside of the
neck zone, which is observed in the neck-like waisted zone of laser-GMA hybrid
welding as well as laser welding;
5. Laser-GMA hybrid welding alters the temperature distribution of the weld fusion zone
and substantially improves the cooling rate near the weld bead neck zone; the
magnitude of cooling rate is lower than that of laser welding. The cooling rate attains
low value in the region of neck zone due to higher solidification time and increases near
the bottom of the keyhole;
6. The values of strain rate gradient near the neck region of laser welding are on the whole
higher than those of hybrid laser-GMA welding. Laser welding is more susceptible to
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 161
liquation cracking and the crack is located more in the neck region during lower heat
input compared to laser-GMA hybrid welding;
7. The weld pool shape has a strong influence on the stress-strain pattern, and a neck weld
pool shape can have a detrimental effect on the HAZ cracking behavior. Adding the arc
heat source alters the strain distribution, and strain rate gradient provides theoretically
understanding of how laser-GMA hybrid welding produces less susceptibility to
liquation cracking compared to laser welding.
Author details
Zhiguo Gao
University of Manitoba, Canada
Acknowledgement
Thanks a lot to the Journal of Acta Materialia, Journal of Materials Science Research and
International Journal of Advanced Manufacturing Technology, the author makes reference
to some previous works in this chapter.
4. References
Akbari, D. & Sattari-Far, I. (2009). Effect of the welding heat input on residual stresses in
butt-welds of dissimilar pipe joints. International journal of pressure vessels and piping,
Vol.86, No.11, pp. 769-776
Anderson, T.; DuPont, J. & DebRoy, T. (2010). Origin of Stray Grain Formation in single-
crystal superalloy weld pools from Heat Transfer and Fluid Flow Modeling. Acta
Materialia, Vol. 58, No.4, pp.1441-1454
Banerjee, P. & Overfelt, R. (1999). Viscosity measurements of industrial alloys using the
oscillating cup technique. International Journal of Thermophysics, Vol.20, No.6, pp.1791-
1800
Bonifaz, E. & Richards, N. (2009). Modeling cast IN-738 superalloy gas tungsten arc welds.
Acta Materialia, Vol.57, No.57, pp.1785-1794
Brooks, R.; Monaghan, B. & Barnicoat A. (1996). The Physical properties of alloys in the
liquid and Mushy states, International Journal of Thermophysics, Vol.17, No.5, pp.1151-
1161
Bussac, A. & Gandin, C. (1997).Prediction of a process window for the investment casting of
dendritic single crystals. Materials Science and Engineering: A, Vol.237, No.1, pp. 35-42
Chang, K. & Lee, C. (2009). Finite element analysis of the residual stresses in T-joint fillet
welds made of similar and dissimilar steels. Int J Adv Manuf Technol, Vol.41, No.3-4,
pp.250-258
Claus,B.&Flemming,O.(2005). Review of laser hybrid welding. Journal of Laser
Applications,Vol.17,No.1,pp.2-13

Welding Processes 162
Dai, K. & Shaw, L. (2001). Thermal and stress modeling of multi-material laser processing.
Acta Materialia, Vol.49, No.20, pp. 4171-4181
Dong, H.(2007). Analysis of Grain Selection during Directional Solidification of Gas Turbine
Blades, Proceedings of the Word Congress on Engineering 2007, Vol.II WCE 2007, ISBN: 978-
988-98671-2-6, London, U.K., July 2-4, 2007
Egbewande, A.; Buckson, R. & Ojo, O. (2010). Analysis of laser beam weldability of Inconel
738 superalloy, Materials characterization, Vol.61, No.5, pp.569-574
Farzadi, A.; Serajzadeh, S. & Kokabi, A. (2010) Investigation of weld pool in aluminum
alloys: Geometry and solidification microstructure. International journal of thermal
sciences, Vol. 49, No.5, pp.809-819
Gao, Z.; Wu, Y. & Huang, J. (2009). Analysis of weld pool dynamic during stationary laser-
MIG hybrid welding. Int J Adv Manuf Technol, Vol.44, No.9-10, pp.870-879
Gao, Z. (2012).Numerical modeling to understand liquation cracking propensity during
laser and laser hybrid welding (I).Int J Adv Manuf Technol, Doi:10.1007/s00170-012-3907-
8,pp.1-13
Gao, Z. & Ojo, O.(2012a) Modeling analysis of hybrid laser-arc welding of single-crystal
nickel-base superalloys, Acta Materialia ,Vol.60,No.6-7,pp.3153-3167
Gao, Z. & Ojo, O.(2012b) Numerical modeling of HAZ liquation cracking tendency during
laser and hybrid laser-arc welding processes, Journal of Materials Science Research ,
Vol.1,No.2,pp.42-55
Gumann, M.; Bezenon, C.; Canalis, P. & Kurz, W. (2001).Single-crystal laser deposition of
superalloys: processing microstructure maps. Acta Materialia, Vol.49, No.6, pp.1051-1062
Grong, . (1997). Metallurgical Modeling of Welding, The Institute of Materials, London, U.K.
He, X.; Elmer, J. & DebRoy, T. (2005). Heat transfer and fluid flow in laser microwelding. J.
Appl. Phys.Vol. 97, No.8, pp.084909-084909-9
Hee, S.; Han, S.; You, C. & Sung, M. (2010). Analysis of residual stress on AH32 butt joint by
hybrid CO2 laser-GMA welding. Computational materials science, Vol. 49, No.2, pp.217-
221
Hong, J.; Park, J.; Park, N.; Eom, I.; Kim, M. & Kang, C. (2008). Microstructures and
mechanical properties of Inconel 718 welds by CO2 laser welding. Journal of materials
processing technology, Vol.201, No.1-3, pp.515-520
Hu, B. & Richardson, I. (2006). Mechanism and possible solution for transverse solidification
cracking in laser welding of high strength aluminium alloys. Materials science and
engineering, A ,Vol.429,No.1-2,pp.287-294
Hunt J. (1984). Steady state columnar and equiaxed growth of dendrites and eutectic.
Materials Science and Engineering, Vol.65, No.1, pp.75-83
Hunziker, O.; Dye, D. & Reed, R. (2000). On the formation of a centerline grain boundary
during fusion welding. Acta Materialia, Vol.48, No.17, pp. 4191-4201
Kong, F.& Kovacevic, R. (2010). 3D finite element modeling of the thermally induced
residual stress in the hybrid laser/arc welding of lap joint. Journal of Materials Processing
Technology, Vol.210, No.6-7, pp.941-950
Liu, W. & DuPont. J. (2004). Effects of melt-pool geometry on crystal growth and
microstructure development in laser surface-melted superalloy single crystals.
Numerical Modelling to Understand Cracking Phenomena
During Laser-GMA Hybrid Welding Nickel-Base Superalloys 163
Mathematical modeling of single-crystal growth in a melt pool (part I). Acta Materialia,
Vol.52, No.16, pp.4833-4847
Liu, W. & DuPont. J. (2005). Effects of substrate crystallographic orientations on crystal
growth and microstructure development in laser surface-melted superalloy single
crystals. Mathematical modeling of single-crystal growth in a melt pool (Part II). Acta
Materialia, Vol.53, No.5, pp.1545-1558
Liu, F.; Lin, X.; Yang, G.; Song, M.; Chen, J.& Huang, W.(2011). Microstructure and residual
stress of laser rapid formed Inconel 718 nickel-base superalloy. Optics & laser technology,
Vol.43, No.1, pp.208-213
Long, R.; Liu, W.; Xing, F. & Wang H. (2008). Numerical simulation of thermal behavior
during laser metal deposition shaping. Transactions of Nonferrous Metals Society of China,
Vo.18, No.3, pp.691-699
Luo, X.; Shinozaki, K. & Kuroki, H. (2002). Analysis of temperature and elevated
temperature plastic strain distributions in laser welding HAZ study of laser weldability
of Ni-base superalloys. Welding International, Vol.16, No.5, pp.385-392
Moore, P.; Howse, D. & Wallach, E. (2004). Microstructures and properties of laser/arc
hybrid welds and autogenous laser welds in pipeline steels .Sci Technol Weld Joining,
Vol. 9, No.4, pp.314-322
Nishimoto, K.; Woo, I. & Shirai, M. (2002). Analyses of temperature and strain distributions
in laser welds study of the weldability of Inconel 718 cast alloy. Welding International,
Vol.16, No.4, pp.284-292
Park, J.; Babu, S.; Vitek, J.; Kenik, E. & David, S. (2003). Stray grain formation in single
crystal Ni-base superalloy welds. Journal of Applied Physics, Vol.94, No.6, pp.4203-4209
Pottlacher, G.; Hosaeus, H.; Kaschnitz, E. & Seifter, A.( 2002) Thermophysical properties of
solid and liquid inconel 718 alloy. Scandinavian Journal of Metallurgy, Vol.31, No.3,
pp.161-168
Rai, R.; Roy, G& DebRoy, T. (2007). A Computationally Efficient Model of Convective Heat
Transfer and Solidification Characteristics during Keyhole Mode Laser Welding. J Appl
Phys.,Vol. 101,No. 05409,pp.1-11
Ral, R.; Palmer, T.; Elmer, J. & Debroy, T. (2009). Heat Transfer and Fluid Flow during
Electron Beam Welding of 304L Stainless Steel Alloy. Welding Journal, Vol. 88, No.3,
pp.54-61
Rappaz, M.; David, S.; Vitek, J. & Boatner, L. (1990). Analysis of solidification
microstructures in Fe-Ni-Cr Single-Crystal welds. Metallurgical Transaction A, Vol.21A,
No.6, pp.1767-1782
Ribic, B.; Palmer,T. & DebRoy,T. (2009). Problems and issues in laser-arc hybrid welding.
International Materials Reviews, Vol.54, No.4, pp.223-244
Roy, G.; Elmer, J. & DebRoy, T. (2006). Mathematical modeling of heat transfer, fluid flow,
and solidification during linear welding with a pulsed laser beam. J. Appl. Phys.Vol.100,
No.3, pp.034903-034903-7
Stelling, K.; Boellinghaus,Th.; Lammers,M. & Schobbert, H. (2005). Laser plasma power
hybrid welding in vertical-up and vertical-down positions, Proceedings of the 7th

Welding Processes 164
international conference on trends in welding research, pp. 121-126,Callaway gardens resort,
Pine mountains, Georgia, USA, May 16-20,2005
Vitek. J. (2005). The effect of welding conditions on stray grain formation in single crystal
welds-theoretical analysis. Acta Materialia, Vol. 53, No.1, pp., 53- 67
Wang, N.; Mokadem, S.; Rappaz, M. & Kurz, W. (2004). Solidification cracking of superalloy
single- and bi-crystals. Acta materialia, Vol.52, No.1, pp.3173-3182
Wu, S.; Zhao, H.; Wang, Y. & Zhang, X. (2004). A new heat source model in numerical
simulation of high energy beam welding. Transactions of the china welding institution,
Vol.25, No.1, pp.91-94
Yann, D.; Eric, L. & Corinne, A. (2010). Numerical modeling of Inconel 738LC deposition
welding: prediction of residual stress induced cracking. Journal of Materials Processing
Technology, Vol.210, No.14, pp.2053-2061
Yilbas, B.; Akhtar, S. & Karatas, C. (2010). Laser surface treatment of Inconel 718 alloy:
Thermal stress analysis. Optics and Lasers in Engineering, Vol.48, No.7-8, pp.740-749
Zeng, D. (2006). Annular beam shaping and optical trepanning, In: Doctor of Philosophy,
22.02.2012, Available from
http://etd.fcla.edu/CF/CFE0001333/Zeng_Danyong_200612_PHD.pdf
Chapter 8




2012 Kong and Kovacevic, licensee InTech. This is an open access chapter distributed under the terms of
the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Development of a Comprehensive Process
Model for Hybrid Laser-Arc Welding
Fanrong Kong and Radovan Kovacevic
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/45850
1. Introduction
Recently, a hybrid welding technique combining laser welding and arc welding has been
finding broader applications in industry due to its unique advantages, such as higher
welding efficiency and lower costs [1, 2] compared to the traditional arc welding or
autogenous laser welding. Because of rapid melting and solidification occurring in the weld
zone, a locally high thermal gradient inevitably exists and accompanies with the whole
welding process, which really decides the final residual stress and distortion distributions of
weld and affects the remained grain size in the fusion zone (FZ) and heat affected zone
(HAZ). Also, the levels of residual stresses and distortions directly influence the weld
quality [3]. In comparison with traditional arc welding and autogenous laser welding, the
temperature field and residual stress distribution in hybrid laser-arc welding involve more
variables because of the additional interaction between the laser and arc plasma [4] thus
becoming much more complex and difficult to theoretically and experimentally analyze.
Trial-and-error experiments are not able to fully describe those physical mechanisms
involved in the hybrid laser-arc welding process. Therefore, the numerical tools have been
widely used to help explain the complex welding mechanisms present in the hybrid laser-
arc welding process [5, 6].
Up to now, numerical work on the welding process mostly concentrates on traditional
electric arc welding, including gas tungsten arc welding (GTAW) [7 - 9], submerged arc
welding (SAW) [10,11], gas metal arc welding (GMAW) [12-16], and partly on laser beam
welding (LBW) [17-21]. These studies focused on the heat and mass transfer phenomena in
the weld pool [8, 11-15], thermal-induced distortion and residual stresses [16, 17, 19, 20],
solidification-induced dendrite growth in the FZ [21], and recrystallization in the heat
affected zone (HAZ) [9]. Due to the locally rapid melting and solidification occurring in the
welding process, a high temperature gradientwhich inevitably exists in the weld zone

Welding Processes 166
causes a high-stress concentration in the weld zone and nearby HAZ [22], which usually
exceeds the yield strength of the material. Large residual stresses presented in the welded
structure can obviously reduce the fatigue strength of metal components, causing crack
generation and shorten the lifetime of metal component [23], which could possibly have
disastrous results. A number of mitigation procedures to reduce and/or eliminate the level
of residual stress have been presented by researchers, mainly including enhancement of the
material ductility of solidification zone (SZ) and HAZ, and improving the thermal and
mechanical conditions in the welding processes [24].
Because of the complex physical mechanisms in the welding processwhich are related to
the heat source properties, material performance, and welding parameters, etc.trial-and-
error methods to optimize welding parameters takes a long time and is usually more costly.
Also, understanding of the physics of the welding process is limited by only using an
experimental approach. Numerical simulation as accompanied by theoretical analysis has
been widely applied as a cost-efficient way to help explore the welding phenomena in
different welding techniques. Eagar et al. [25, 26] spent a lot of time in developing
theoretical models for GTAW processes. Dong et al. [27-29] developed numerical models to
predict the residual stresses as well as fatigue life of weld obtained by the multi-pass
welding process. Deng et al. [30-32] developed a series of numerical models to study the
residual stress distribution in variable welding joints.
Compared to the traditional electric arc welding, laser welding has unique advantages such
as high energy density, narrow HAZ, low heat input, and high energy efficiency. However,
laser welding is also limited by its disadvantages like poor gap bridgeability and high
equipment cost. In order to fully use the advantages of both laser and arc welding
techniques, Steen et al. [33] introduced for the first time a hybrid technique by combining
the laser beam and arc for welding and cutting in the late 1970s. Subsequently, researcher
and engineers have presented a number of works on combining the laser and electric arc in
the past decades. Considering that interaction between laser beam and arc plasma is
complex, the hybrid laser-arc welding and cladding processes have not been understood
fully. Most available literature on these approaches is limited at the level of the experimental
study including hybrid laser-GTAW, hybrid laser-GMAW, and hybrid laser-plasma arc
welding for steels, magnesium alloy, aluminum alloy, titanium alloy and dissimilar
materials. In order to further study the welding mechanism of hybrid laser and arc, it is
necessary to develop a comprehensive model to understand the heat and mass transfer,
residual stress evolution, as well as microstructure formation in the hybrid laser-arc welding
process. Zhou and Tsai [34, 35] presented heat transfer and fluid flow models to study the
metal inert gas (MIG) welding and laser-MIG hybrid welding processes. Rao et al. [36]
reviewed the modeling of hybrid laser-gas metal arc (GMA) welding and presented further
studies on synergistic interaction between the laser beam and arc, the metal transfer
features, and behavior of shielding gas. Ribic et al. [37] developed a three-dimensional (3-D)
finite volume model to study heat transfer and fluid flow in the hybrid laser-GTA welding
process. Considering that the microstructure formation of weld has a close relationship with
the macro-scale heat transfer and fluid flow, and residual stress fields, it will be very

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 167
necessary to integrate the thermal, fluid flow and mechanical modeling with the
microstructure evolution like grain growth in the fusion zone and HAZ. Multi-scale and
multi-physics modeling is one of most interesting simulation trends in the laser-based heat
processes, especially in the hybrid laser-arc welding process.

Figure 1. Schematic view of hybrid laser-GMA welding system
In this chapter, a 3-D mathematical model will be developed to numerically predict the
transient temperature distributions and residual stresses in the hybrid laser-GMA welding
of a thick plate of A514 steel in butt joint configuration, as shown in Figure 1. The numerical
solution is achieved based on a finite element method by using a commercial numerical
package, ANSYS. A Monte Carlo model is introduced to consider the grain growth and
phase transformations in the HAZ. The laser and arc heat inputs and heat losses at the
surface of coupons are considered by using ANSYS Parametric Designed Language (APDL).
The influences of the processing parameters (including welding speed, laser power, wire
feed rate, arc power, and stand-off distance from laser to arc) on the profile and geometrical
size of the molten pool, residual stress distribution of the weld, and grain size in the HAZ
are numerically studied. The numerically obtained results are experimentally verified.
2. Finite element modeling
2.1. Thermal analysis of hybrid laser-GMA welding
In the hybrid laser-GMA welding process, laser and GMA simultaneously heat the coupon
surface in local area, which makes the thermal distribution of weld much more complex. In
this study, a cylindrical volume heat-source model with a Gaussian distribution is assumed
to simulate the heat input by laser, and a double-ellipsoidal volume heat source is selected
to consider the heat input by GMA welding. The general thermal governing equation is
shown below, in which thermal conduction-induced heat transfer is considered and
temperature-dependent material properties are used [38].
GMAW Laser







Molten pool
Solidified weld
zone
Welding direction

Welding Processes 168

{ } { }
( )
T
p l a
T
c L D L T q q
t

c
( = + +

c
(1)

0 0
where 0 0 ,
0 0
xx
yy
zz
k
D k
k
(
(
( =
(

(

(2)

{ }
and ,
x
y
z
L
c
c
c
c
c
c



=
`


)
(3)
is the density, cp is the specific heat, T is the temperature, t is time, kxx, kyy, and kzz are the
thermal conductivity components along the x, y, and z axis, respectively;
l
q and
a
q are the
volume heat generation rates due to the laser and GMA heat input, respectively.
So far, a number of heat source models have been developed to simulate the arc welding
and laser welding processes. Laser welding usually consists of laser conduction welding or
laser keyhole welding. The former one has lower energy density as compared with the latter
one by which a keyhole is formed in the weld pool. A surface heat flux model is usually
applied in the thermal analysis of laser conduction welding. However, a volume-distributed
heat source model, like rotary Gaussian heat density distribution [39], is usually used for
simulating a laser keyhole welding. Compared to the laser beam welding, electric arc
welding has much lower energy density, and surface heat flux models with Gaussian
distributions used to be applied to simulate the arc heat input in the arc welding process.
Considering that the enthalpy brought into the weld pool by melted wire in GMAW,
volume-distributed heat source models are preferred, such as hemi-spherical power density
distribution [40], ellipsoidal power density distribution [41], and double ellipsoidal power
density distribution [42]. However, all of these heat source models are empirically derived
based on the experimentally fitting data. Therefore, each heat source model mentioned
above has a certain applicable range in the real production case. It is suggested that
engineers in the welding process design should reasonably select a heat source model which
matches well with the specific welding process. There are a limited number of publications
available to numerically describe the hybrid laser and arc welding process because of lack of
knowledge on complex interaction between the material, arc plasma, and laser beam [43].
Current heat source models of hybrid laser and arc including GTAW and GMAW were
mostly developed with the help of experimental support [34-37].
In this study, a double-ellipsoidal heat source model is introduced to simulate the GMAW
heat input, and a cylindrical heat source model with a sectional Gaussian distribution is
used to consider the laser heat input. q
f
arc (x, y, z, t) and q
r
arc (x, y, z, t), depict heat input
distributions inside the front and rear quadrants of the GMAW heat source, respectively,
which can be expressed as follows [42, 44]:

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 169

2 2 2
0
2 2 2
6 3
3( ) 3( ) 3( )
( , , , ) exp exp exp
f arc
f
w
arc
f f
f P
x x y L z vt
q x y z t
a b c abc t t
| |
| | | |

|
| | =
| |
|
\ . \ .
\ .
(4)

2 2 2
0
2 2 2
6 3 3( ) 3( ) 3( )
( , , , ) exp exp exp
r r arc w
arc
r r
f P x x y L z vt
q x y z t
a b c abc t t
| | | | | |

| | | =
| | |
\ . \ . \ .
(5)
where a, b, cf, cr are the characteristic parameters of heat sources, and a, b, cf, and cr are set at
4 mm, 3 mm, 3 mm and 7 mm respectively [44]. Parc denotes the nominal power of the
GMAW, and Parc=UI. Where is the energy efficiency of GMAW based on the welded
metal, U denotes the arc voltage of GMAW, and I stands for arc current of GMAW.
( , , , )
laser
q x y z t denotes the laser radiation-induced volume heat input given by [45, 46]:

2 2 2
0
2 2
2
cos ( ) [ ( ) tan ] cos
( , , , ) exp
2 2
( / )
laser w la
laser l
l l
w
P x x z L y vt D
q x y z t
R R
y L
q
t
| |
u + + u u
| =
|
\ .

(6)
where l is laser absorption efficiency based on the welded material, Plaser stands for the
nominal power of the laser beam, x0 is the x-coordinate of the center point of laser spot at the
coupon surface, Lw is the thickness of the butt joint, and Rl is the effective radius of the laser
beam, is the inclination angle of laser head, Dla is the laser-to-arc stand-off distance. is
set at 0
o
and Dla is set at 8 mm in this study, and v denotes the welding speed.
The boundary conditions at the sample surfaces are given by:
( )
c
T
k h T T
n

c
=
c
(7)
where n is the normal outward vector to the surface of specimen, T

is the room
temperature, and hc is the heat transfer coefficient of sample surface.
2.2. Mechanical analysis of hybrid laser-GMA welding
The mechanical analyses of hybrid laser and arc welding are similar to the previous studies
on the electric arc welding and laser welding. The stress and distortion of weld are mainly
caused by the thermally-induced expansion and shrinkage and the accompanying phase
transformation-induced volume change. By considering the elastic-plastic material
properties, stress and strain relationships in the hybrid laser-GMA weld are given by [38]:

{ } [ ]{ }
el
e
D o c =
(8)
where {} denotes the stress vector, [De] denotes elastic stiffness matrix, and {
el
}denotes
elastic strain vector expressed by [30]:
{ } { } { } { } { } { }
pl Trp el th V
c c c c c c
A
= (9)

Welding Processes 170
where {} is the total strain vector, {
th
} is thermal strain vector, {
pl
} is the plastic strain
vector, {
V
}

is strain vector due to phase transformed induced volume change, and {
Trp
} is
strain vector due to phase transformation plasticity which is ignored in this study. The
boundary conditions taken into consideration in the mechanical analysis assume that one
edge of the butt joint is fixed, and the other one is only transversely shrinkage free.
2.3. Grain size prediction model by Monte Carlo method
Grain size evolution and phase transformation play a critical role in deciding the final
mechanical properties of weld, and it is necessary to involve those factors in the thermo-
mechanical modeling of different welding processes. Many good trials have been performed
to numerically predict the grain growth in the fusion zone and heat affected zone for
solidification and re-crystallization, respectively, which includes Monte Carlo (MC) model
[47-50], phase field (PF) method [51], and cellular automaton (CA) model [52] combined
with finite element and finite difference analyses. Here a brief introduction of MC model to
predict the grain growth in HAZ will be performed. The detailed description of phase field
method and cellular automaton model-based numerical prediction of grain growth in welds
can be found in literature [53].
Monte Carlo model-based grain growth prediction generally includes the following several
steps: (1) The representation of the considered material in a two-dimensional (2-D) or 3-D of
cells, as shown in Figure 2a. The content of each cell stands for its crystallographic
orientation. A region consisting of a set of consistently distributed cells with the same
orientation value denotes a grain. The grain boundaries are identified by a curve in 2-D
matrix or a surface in 3-D matrix between the separate planes or volumes with different
orientations. (2) After selecting the matrix type and defining it by an initially random
number, the free energy of a cell in the matrix with its specific crystallographic orientation
based on its surroundings will be identified. (3) Randomly selecting a new crystallographic
orientation for each cell. (4) Calculating the free energy of the new coming element with the
new crystallographic orientation, the two energy values and their difference are then
calculated. A new grain orientation that will minimize the free energy is generated with the
selected transition probability [54]. These four steps will be reiterated many times at random
positions in the matrix. The ultimate product is a microscopic simulation of the free energy
decaying in the system, which is in fact the main driving force for grain growth. The
Hamiltonian demonstrates the interaction among the closest neighbors in a particular cell,
which stands for the grain boundary energy and can be calculated as follows [54]:
( 1)
i j
nn
G J s s

(10)
where, J is a positive constant that characterizes the scale of the grain boundary energy;
i
s is
one of the possible orientations, which is set between 1 and q, in the
th
i cell of the matrix;
j
s
is the crystallographic orientation of one neighboring cell; nn is the amount of neighboring
cells for each element. In the Monte Carlo model, a Moore neighborhood model is selected

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 171
(see Figure 2b); therefore, nn=8.
ab
o is the Kronecker-delta, which equals 1 when two
elements in the matrix are equal, i.e., a=b, and 0 for others. As a consequence, neighboring
cells with different orientations contribute a free energy J to the system and 0 with the same
orientation. The total number of grain orientations, q is set at 40 in this model since it is
known that the grain-growth exponent becomes almost independent of q when its value is
larger than 30 [54].

Figure 2. The grain structure in MC model with Moore neighborhood [51]
(a) The grain structure represented by a 2D square (b) Moore neighborhood
The transition probability W is given by [55]:

exp( ) , 0
1, 0
b
G
G
k T W
G
A
A >

A s

(11)
where G is the change of the free energy because of the orientation alteration,
b
k is the
Boltzman constant (kb=1.3810
-23
m
2
kg/s
2
/K), and T is the temperature. Therefore, the speed
of the moving segment can be calculated by [55]:

1
[1 exp( )]
i
b
G
v C
k T
A
= (12)
where C1 is the boundary mobility. For a continuous grain growth, the final grain size can be
calculated by using the following equation [55]:

0
( )
n n
L L f T t = (13)
(a)
(b)

Welding Processes 172
where L and L0 are the final and initial mean grain sizes respectively calculated by the
linear-intercept method, n is the grain growth exponent and set at 1.84 in this study [56]. f(T)
is usually computed as an Arrhenius-type equation [55], and its expression is shown as
follows:
( ) exp( )
g
Q
f T K
R T
(14)
where K is the pre-exponential coefficient, Q is the activation energy for grain growth, and
Rg is the universal gas constant. In this study, K is set at 3.0110
-2
, and Q is set at 1.710
5

J/mol [56].
The Monte Carlo method has been proven to be an effective way to simulate grain growth
with slow and uniform temperature evolution such as metal casting [53]. In the hybrid
welding by laser and arc, there exists a dynamic thermal process with rapid heating and
cooling resulting in an abrupt temperature gradient in the HAZ and fusion zone. In the
simulation of microstructure evolution, three techniquessuch as the atomistic models, a
grain boundary migration (GBM), and experimentally data-based (EDB) modelshave been
presented [53, 58-60]. The atomistic model used to be only applied to small numbers of atoms
like nanocrystals [60], and it is not suitable for a large-scale FZ or HAZ simulation. The GBM
model can be a good alternative for grain-growth simulation when the isothermal grain-
growth kinetics is not accessible. However, the physical properties of the material in this
model have to be known, and the grain size is assumed to be proportional to the square root of
time. The EDB model can avoid these shortages and be applied to simulate the grain growth in
HAZ when the isothermal grain-growth kinetics of metal are available. Therefore, it can be
used to relate time and temperature to the Monte Carlo simulation-time step tMCS [55]:

1
1
( )
n
MCS
L K t (15)
where is the discrete grid-point spacing in the Monte Carlo model, and K1 and n1 are
constants. Through the regression computation of tMCS and the Monte Carlo model
predicting the grain size, the values of K1 and n1 are obtained as 0.715 and 0.477, respectively
[56]. In the EDB model, the relationship between the tMCS and the real time-temperature T(t)
is further given by [60]:

1
0
1
1
( ) ( ) (exp( ) )
( )
( )
nn n
MCS i
n
g
L K Q
t t
K R T t
K

(16)
where n is the grain growth exponent, T(t) is the mean temperature in a time interval ti.
Therefore, at any given monitoring location where the temperature is known as a function
of time, tMCS can be related to the real time t, which is ti. The tMCS values at different
locations calculated through Eq. (16) cannot be straightly applied to the Monte Carlo model
since the selection of a grid point for updating the orientation number is stochastic in the
Monte Carlo approach. Consequently, the probability of choosing each grid point is the

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 173
same as in the traditional MC calculations. However, grains usually grow at higher rates in
the HAZ region of higher temperature, where a sharp temperature gradient is present. This
fact has to be included in any practical grain-growth calculation scheme. One solution is to
develop a scheme in which grain orientations at higher-temperature locations (higher tMCS
locations) are updated with a higher frequency by considering a probability gradient. In
other words, the site-selection probability changes with location. The larger the tMCS at a
location, the higher the corresponding site-selection probability [57, 60]:

MCS MCSMAX
P t / t (17)
where tMCSMAX is the maximum of tMCS in the simulation domain.


Figure 3. Finite element meshes for hybrid laser-GMA weld

Figure 4. Temperature-dependent thermal and mechanical properties of A514 steel

Welding Processes 174



Figure 5. Numerical procedure performed in the thermo-mechanical FE analysis
Build up the CAD model of welds and define
thermal physical properties
Choose the element type and meshing: a finer
mesh locates at the weld zone
Yes
No
Delete the heat input, and coming into the
cooling phase, and solve this time step.
t
i
=t
i
+t,
move into the next
time step
Define the boundary conditions due to the
convection and radiation, load the heat input, and
solve this time step. Is the laser head and arc torch
reaching the end of the weld?
Calculate the phase transformation induced stress
components and loaded into mechanical model as
initial stress
Predict the grain growth by using Monte
Carlo model
Output the numerical results and simulation
is finished
Define the mechanical constraint, loading the
temperature data step by step, and solve the
transient stress and strain
Switch the element type to mechanical analysis,
delete all the thermal boundary conditions, and
chose the reasonable mechanical material model

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 175
2.4. Implementation of numerical procedure by using APDL
An uncoupled finite element thermo-mechanical model with considering the grain growth
in the HAZ by Monte Carlo model is developed to study the temperature distribution and
residual stress field in the hybrid laser-GMA welding process. A non-uniform mesh is
selected in which a finer mesh is used in the weld bead and a course mesh is defined in the
other region of the welded coupons (see Figure 3). A temperature-dependent material
property is used in the numerical modeling, as listed in Figure 4. A thermal FE analysis is
performed to achieve the temperature field of hybrid laser-GMA welding process. The wire
feeding into the groove to form the weld bead has been simulated by using element kill-
and-birth approach which is available in ANSYS software. The achieved geometrical size of
the weld zone could be compared to the micrographs of the weld cross-section obtained by
an optical microscope, by which the accuracy of thermal analysis can be verified. The
numerical model is then transferred to mechanical analysis module in ANSYS by switching
the element type from thermal to structural. The corresponding constraints are exerted into
sample boundaries. The achieved temperature histories are subsequently loaded into the
mechanical model step by step to calculate the displacement, stress/strain of the sample due
to the thermal expansion or shrinkage during the welding process. A bilinear hardening
principle is introduced in this study to simulate the material plastic behavior. Von Mises
criterion is used for considering the yield behavior of the sample material. Figure 5 shows
the numerical procedure used in this study.
3. Experimental set-up
A 4 kW fiber laser and a GMAW torch are mounted on a robotic arm to perform the hybrid
welding of a thick plate for a butt joint configuration. The photo of an experimental set-up
for welding is shown in Figure 6. In order to control the gap thickness, spot welding is
performed at both ends of the joint before the formal welding starts. After the welding
process is completed, the achieved sample will be cut into standard tensile coupon for
tensile test, the left parts will be mounted for polishing, and etching to test the micro-
hardness and microstructure. The base metal is high strength steel A514; its chemical
composition is listed in Table 1. The wire material is ER100S-G. Its diameter is 0.9 mm, and
its chemical composition is listed in Table 2. Residual stresses were measured by using the
X-ray diffraction technique. Before performing the residual stress measurement, the
measurement areas were cleaned by using polishing paper.
C Mn P S Si Cr Mo V Ti B
Min. 0.12 0.70 0.20 0.40 0.15 0.03 0.01 0.0005
Max. 0.21 1.00 0.035 0.008* 0.35 0.65 0.25 0.08 0.04 0.005
Table 1. Chemical composition of A514 [61]
Cu % max Ni % max Fe % max Mn % max Mo % max
<0.5 <5.0 Balance <5.0 0.50
Table 2. Chemical composition of ER100S-G [62]

Welding Processes 176

Figure 6. Photo of hybrid laser-GMA welding system used in this study
4. Results and discussion
4.1. Experimental results
Figure 7 shows the surface morphology of A514 weld obtained by hybrid laser-GMA
welding, and Figure 8 presents the corresponding cross-sectional view of weld. It can be

Figure 7. A514 sample achieved by 3.8 kW laser and 159A30.5V GMAW with a welding speed of 12
mm/s and 8-mm stand-off distance between the laser and arc

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 177
seen that a sound weld quality is achieved by using hybrid laser-GMA welding and the
welding-induced cracks can be effectively mitigated by reasonably selecting filler wire
matched with base metal. Also, a sound mechanical property can be obtained. Figures 9a
and b show the hardness distribution in the weld obtained by hybrid laser-GMA
welding.



Figure 8. Cross-sectional view of A514 weld sample achieved by 3.8 kW laser and 159A30.5V GMAW
with a welding speed of 12 mm/s and 8-mm stand-off distance between the laser and arc
4.2. Thermal results and grain size prediction in the HAZ of hybrid laser-GMA
welding
Finite element analyses results show the temperature at the each location of weld with
respect to the welding time. The temperature evolution curves at position A, B, and C are
shown in Figure 10, where position A is located at the center of weld, position B is at the
heat-affected zone, and position C is in the base metal, as shown in Figure 11b. Figures 11a
and b show the top and cross-sectional views of weld obtained by hybrid laser-GMA
welding, respectively. It is inevitable that material heating and cooling is accompanied by
phase transformation and grain size change, especially in the HAZ of weld, which is the
weakest zone of the weld. In this study, a Monte Carlo-based sub-model is introduced to
numerically predict the grain growth in the HAZ combined with the finite element thermal
analysis. Figure 12a shows the relationship of temperature versus Monte Carlo step at
Position B, and Figure 12b presents the curve of Monte Carlo step versus real time at
Position B. The corresponding predicted grain size distribution in the Position B is shown in
Figure 12c. The numerically predicted grain size is compared to grain size shown in the
micrograph of the cross-section of weld (see Figure 12d), and a qualitative agreement is
achieved.

Welding Processes 178









Figure 9. Hardness distribution transverse to the weld in A514 sample, (a) at the top surface and (b) at
the bottom surface

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 179




Figure 10. Temperature evolution curve at the FZ, HAZ and BM during the hybrid laser-GMA welding
of A514 steel




Figure 11. Numerically predicted isotherms at the top (a) and at the cross-section I-I of the weld (b)
obtained by hybrid laser-GMA welding
C B A

Welding Processes 180

Figure 12. (a) Temperature versus Monte Carlo step, (b) Monte Carlo step versus real time, (c)
numerical predicted and (d) experimentally measured grain size distributions at Position B in the HAZ
of hybrid A514 weld by Monte Carlo sub-model
4.3. Thermally-induced residual stress in the hybrid laser-GMA welding
A finite element analysis is further performed to predict the thermally-induced residual
stress distribution based on the previous thermal analysis results. The contours of transient
stress, along thickness normal stress, longitudinal stress and equivalent residual stress of
hybrid weld are shown in Figures 13a through d, respectively. It can be seen that the higher
stress concentrations are located at the weld zone, which also indirectly verifies the
previously experimentally obtained conclusions that the thermally induced cracks are
usually generated at the weld zone, not in the base metal. The corresponding contours of
stress distribution of the cross-section in the middle of weld length are shown in Figures 14a
through d. A higher stress concentration is found to be located at the top region of cross-
section
Figures 15a through c show residual stress distribution transverse to the weld bead at the
different thicknesses in the middle of weld obtained by hybrid laser-GMA welding. It also
validates the conclusion driven from Figure 14 that high tensile transverse and longitudinal
stresses are located at the top and bottom regions of the weld center, high compressive
transverse stresses are located at a half of the weld thickness. From the equivalent stress
distribution point of view, the peak value of stress concentration is a little lower than that at
the top and bottom of the weld. Figure 16 shows residual stress distribution transverse to
the weld bead at the different locations along the top surface of weld obtained by hybrid
laser-GMA welding. Figure 17 also shows residual stress distributions along the central line

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 181

Figure 13. (a) Transverse stress SX, (b) along-thickness normal stress SY, (c) longitudinal stress SZ, and
(d) von Mises equivalent residual stress SEQV mapping of weld by hybrid laser-GMA welding (unit of
stress in the contour is Pa)

Figure 14. (a) Transverse stress SX, (b) along-thickness normal stress SY, (c) longitudinal stress SZ, and
(d) von Mises equivalent residual stress SEQV mapping of cross-section of weld by hybrid laser-GMA
welding (unit of stress in the contour is Pa)

Welding Processes 182
at the top surface of weld centerline achieved by hybrid laser-GMA welding. It is clear that
stress distribution across the weld bead is uniform along the weld; only a little drop in stress
magnitude exists at the both ends of weld.




Figure 15. Residual stress distribution transverse to the weld bead at the different thicknesses in the
middle of weld (z=30 mm) obtained by hybrid laser-GMA welding
Figure 18 shows a comparison of experimentally-measured and FE numerically-predicted
residual stress distributions at the middle of weld length of top surface of weld by hybrid
laser-GMA welding. There is a qualitative agreement between the developed numerical
model and experimentally measured stress by an X-ray diffraction technique. Figures 19, 20
and 21 show the transverse, longitudinal, and equivalent stresses as well as temperature
evolution with time at the positions A, B and C, respectively. It can be seen that the peak
values of transient stresses at positions A, B, and C are sensitive to the temperature curve at
the position A but not at the positions B and C.
(a) (b)
(c)

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 183

Figure 16. Residual stress distribution transverse to the weld bead at the different locations along the
top surface of weld obtained by hybrid laser-GMA welding

Figure 17. Residual stress distributions along the longitudinal direction of weld at the top surface of
weld centerline achieved by hybrid laser-GMA welding
(a) (b)
(c)

Welding Processes 184

Figure 18. Comparison of experimentally-measured and FE numerically-predicted residual stress
distribution at the middle of weld length of top surface of weld by hybrid laser-GMA welding




Figure 19. Stress and temperature evolution curves at the Position A in FZ of hybrid A514 weld
(a)
(b)
(a)
(b)
(c)

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 185

Figure 20. Stress and temperature evolution curves at the Position B in HAZ of hybrid A514 weld

Figure 21. Stress and temperature evolution curves at the Position C in BM of hybrid A514 weld
(a) (b)
(c)
(a)
(b)
(c)

Welding Processes 186
5. Conclusions
The hybrid approach combining laser and arc has unique features which definitely help to
achieve a better weld quality and to improve the production efficiency. A brief overview of
modeling of hybrid laser-arc welding process has been presented in which heat transfer,
fluid flow, residual stress and distortion, as well as phase transformation in the weld zone
and heat affected zone, are involved. As a case study, a 3D thermo-mechanical finite element
model is developed to study the thermally-induced residual stress in the hybrid laser-GMA
welding process. A Monte Carlo model is introduced to numerically predict the grain
growth in the heat affected zone of weld combined with finite element thermal analysis,
which can be used to further understand the welding mechanisms of hybrid laser-GMA
welding as well as other welding technology.
Author details
Fanrong Kong and Radovan Kovacevic
*

Research Center for Advanced Manufacturing, Lyle School of Engineering,
Southern Methodist University, Dallas, TX, USA
Acknowledgement
The authors would like to thank Ph.D. candidate, Mr. Junjie Ma for his valuable help in the
experiment and technical discussions and Research Engineer, Mr. Andrew Socha for his
support in the experimental set-up. This work was financially support by NSFs Grant No.
IIP-1034652
6. References
[1] C. Roepke, S. Liu, S. Kelly, and R. Martukanitz. Hybrid laser arc welding process
evaluation on DH36 and EH36 steel. Welding Journal, 2010, 89: 140s-150s.
[2] C. Bagger and F.W. Olsen. Review of laser hybrid welding. Journal of Laser
Applications. 2005, 17(1): 2-14.
[3] M. N. James, D. J. Hughes, Z Chen, H Lombard, D. G. Hattingh, D. Asquith, J. R. Yates
and P. J. Webster, Residual stresses and fatigue performance, Engineering Failure
Analysis, 2007,14: 384-385.
[4] P. Kah. Usability of laser-arc hybrid welding processes in industrial applications. Ph.D.
Dissertation, Lappeenranta University of Technology, 2011.
[5] G. Turichin, E. Valdaytseva, I. Tzibulsky, A. Lopota, O. Velichko. Simulation and
technology of hybrid welding of thick steel parts with high power fiber laser. Physics
Procedia, 2011, 12: 646-655.
[6] M.H. Cho and D.F. Farson. Simulation study of a hybrid process for the prevention of
weld bead hump formation. Welding Journal, 2007, 86: 174s-179s.

*
Corresponding Author

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 187
[7] F.G. Lu, S. Yao, S.N. Lou, Y.B. Li. Modeling and finite element analysis on GTAW arc
and weld pool. Computational Materials Science, 2004, 29:371-378.
[8] H.G. Fan, H.L. Tsai, S.J. Na. Heat transfer and fluid flow in a partially or fully penetrated
weld pool in gas tungsten arc welding. International Journal of Heat and Mass Transfer,
2001, 44: 417-428.
[9] H.J. Aval, S. Serajzadeh, A.H. Kokabi. Prediction of grain growth behavior in HAZ
during gas tungsten arc welding of 304 stainless steel. Journal of Materials Engineering
and Performance, 2009, 18: 1193-1200.
[10] L.J. Yang, M.J. Bibby, R.S. Chandel. Linear regression equations for modeling the
submerged arc welding process. Journal of Materials Processing Technology, 1993, 39(1-
2): 33-42.
[11] D. V. Kiran, B. Basu, A.K. Shah, S. Mishra, and A. De. Three-dimensional heat transfer
analysis of two wire tandem submerged arc welding. ISIJ International, 2011, 51 (5):
793-798.
[12] J. Haidar. A theoretical model for gas metal arc welding and gas tungsten arc welding I.
Journal of Applied Physics, 1998, 84: 3518-3529.
[13] J. Haidar. Prediction of metal droplet formation in gas metal arc welding II. Journal of
Applied Physics, 1998, 84: 3530-3540.
[14] J. Haidar. An analysis of heat transfer and fume production in gas metal arc welding.
III, Journal of Applied Physics, 1999, 85 3448-3459.
[15] Z.H. Rao, J. Hu, S.M. Liao, H.L. Tsai. Modeling of the transport phenomena in GMAW
using argon-helium mixtures. Part I- The arc. International Journal of Heat and Mass
Transfer, 2010, 53: 5707-5721.
[16] C. Heinze, C. Schwenk, M. Rethmeier. Numerical calculation of residual stress
development of multi-pass gas metal arc welding. Journal of Constructional Steel
Research, 2012, 72: 12-19.
[17] A. Kaplan. A model of deep penetration laser welding based on calculation of the
keyhole profile. Journal of Physics D: Applied Physics, 1994, 27(9): 1805-1814.
[18] R. Spina, L. Tricarico, G. Basile, T. Sibillano. Thermo-mechanical modeling of laser
welding of AA5083 sheets. Journal of Materials Processing Technology, 2007, 191: 215-
219.
[19] A.P. Mackwood, R.C. Crafer. Thermal modeling of laser welding and related processes:
a literature review. Optics & Laser Technology, 2005, 37: 99-115.
[20] P. Martinson, S. Daneshpour, M. Koak, S. Riekehr, P. Staron. Residual stress analysis of
laser spot welding of steel sheets. Materials & Design, 2009, 30(9): 3351-3359.
[21] W. Tan, N.S. Bailey, Y.C. Shin. A novel integrated model combining cellular automata
and phase field methods for microstructure evolution during solidification of multi-
component and multi-phase alloys. Computational Materials Science, 2011, 50: 2573-
2585.
[22] K. Asim, J. Lee, and J. Pan. Failure mode of laser welds in lap-shear specimens of high
strength low alloy (HSLA) steel sheets. Fatigue & Fracture of Engineering Materials &
Structures. 2011, 35: 219-236.

Welding Processes 188
[23] Y. Arai, M. Kikuchi, T. Watanabe, M. Nakagaki. Residual stress due to welding and its
effect on the assessment of cracks near the weld interface. International Journal of
Pressure Vessels and Piping. 1995, 63(3): 237-248.
[24] Y.P. Yang, P. Dong, J. Zhang and X. Tian. A hot-cracking mitigation technique for
welding high-strength aluminum alloy. Welding Journal, 2000, 79(1): 9s-17s.
[25] A.H. Dilawari, T.W. Eagar and J. Szekely. An analysis of heat and fluid flow
phenomena in electroslag welding. Welding Journal, 1978, 57(1): 24s-30s.
[26] L. A. Jones, T. W. Eager, J. H. Lang. Magnetic forces acting on molten drops in gas metal
arc welding, Journal of Physics D: Applied Physics, 1998, 31: 93-106.
[27] P. Dong. Residual stress analyses of a multi-pass girth weld: 3-D special shell versus
axisymmetric models. ASME Journal of Pressure Vessel Technology, 2001, 123: 207-213.
[28] P. Dong, J.K. Hong, and P. Rogers. Analysis of residual stresses in Al-Li repair welds
and mitigation techniques, Welding Journal (London), 1998, 77 (11): 439-445.
[29] P. Dong, J.K. Hong, J. Zhang, P. Rogers, J. Bynum, and S. Shah, Effects of repair weld
residual stresses on wide-panel specimens loaded in tension. ASME Journal of Pressure
Vessel Technology, 1998, 120: 122128.
[30] Dean Deng, Yu Luo, Hisashi Serizawa, Masakazu Shibahara, and Hidekazu Murakawa.
Numerical simulation of residual stress and deformation considering phase
transformation effect. Transactions of Joining and Welding Research Institute (Osaka
University) 2003, 32(2): 325-333.
[31] D. Deng, H. Murakawa. Numerical simulation of temperature field and residual stress
in multi-pass welds in stainless steel pipe and comparison with experimental
measurements. Computational Materials Science, 2006, 37: 269-277.
[32] D. Deng. FEM prediction of welding residual stress and distortion in carbon steel
considering phase transformation effects. Materials and Design, 2009, 30: 359-366.
[33] W.M. Steen, and M. Eboo. Arc augmented laser welding. Metal Construction, 1979,
11(7): 332-335.
[34] J. Zhou, H. L. Tsai. Modeling of transport phenomena in hybrid laser-MIG keyhole
welding. International Journal of Heat and Mass Transfer, 2008, 51: 4353-4366.
[35] J. Hu, H.L. Tsai. Heat and mass transfer in gas metal arc welding. Part I: The arc.
International Journal of Heat and Mass Transfer, 2007, 50 (5-6): 833-846.
[36] Z.H. Rao, S.M. Liao, and H.L. Tsai. Modelling of hybrid laser-GMA welding: review
and challenges. Science and Technology of Welding and Joining, 2011, 16 (4): 300-305.
[37] B. Ribic, R. Rai, and T. DebRoy. Numerical simulation of heat transfer and fluid flow in
GTA/laser hybrid welding. Science and Technology of Welding and Joining. 2008, 13
(9): 683-693.
[38] ANSYS Inc., ANSYS 11.0 Manual, 2007.
[39] S. Wu, H. Zhao, Y. Wang, X. Zhang. A new heat source model in numerical simulation
of high energy beam welding. Transaction of China Welding Institute, 2004, 25: 9194.
[40] V. Pavelic, R. Tanbakuchi, O.A. Uyehara, and P.S. Myers. Experimental and computed
temperature histories in gas tungsten arc welding of thin plate, Welding Journal
Research Supplement, 1969, 48: 259s-305s.

Development of a Comprehensive Process Model for Hybrid Laser-Arc Welding 189
[41] E.A. Bonifaz. Finite element analysis of heat flow in single-pass arc welds. Welding
research supplement, 2000, May: 121s-125s.
[42] J. Goldak, A. Chakravarti, and M. Bibby. A new finite element model for heat sources,
Metallurgical Transaction B, 1984, 15B: 299-305.
[43] P. Peyffarth, and I.V. Krivt. Welding and Allied Processes Volume I. Laser-Arc
Processes and Their Application in Welding and Material Treatment. Tailor&Francis,
London, 2002.
[44] F. Kong, J. Ma, R. Kovacevic. Numerical and experimental study of thermally induced
residual stress in the hybrid laserGMA welding process. Journal of Materials
Processing Technology, 2011, 211(6): 1102-1111.
[45] C.S. Wu, Q.X. Hu, J.Q. Gao. An adaptive heat source model for finite element analysis
of keyhole plasma arc welding. Computational Materials Science. 2009, 46: 167-172.
[46] F. Kong, R. Kovacevic. 3D finite element modeling of the thermally induced residual
stress in the hybrid laser/arc welding of lap joint. Journal of Materials Processing
Technology, 2010, 210 (6-7): 941-950.
[47] M. Nosonovsky, X. Zhang, and S.K. Esche. Scaling of Monte Carlo simulation of grain
growth in metals. Modelling and Simulation in Materials Science and Engineering,
2009, 17: 025004 (13pp).
[48] S.Mishra, T.Debroy. Measurements and Monte Carlo simulation of grain growth in the
heat-affected zone of Ti-6Al-4V welds. Acta Materialia 2004, 52: 1183-1192.
[49] S.Sista, Z.Yang, and T. Debroy. Three-dimensional Monte Carlo simulation of grain
growth in the heat-affected zone of a 2.25Cr-1Mo steel weld. Metallurgical and
Materials Transactions B, 2000, Volume 31B: 529-536.
[50] F. Kong, S. Santhanakrishnan, D. Lin, and R. Kovacevic. Modeling of temperature field
and grain growth of a dual-phase steel DP980 in direct diode laser heat treatment.
Journal of Materials Processing Technology, 2009, 209 (18-19): 5996-6003.
[51] A. Farzadi, M Do-Quang, S. Serajzadeh, A.H. Kokabi, and G. Amberg. Phase field
simulation of weld solidification microstructure in an Al-Cu alloy. Modelling and
Simulation in Materials Science and Engineering, 2008, 16: 165005 (18pp).
[52] Ch.A. Grandin, J.L. Desbiolles, M. Rappaz, and Ph. Thevoz. A three-dimensional
cellular automaton-finite element model for the prediction of solidification grain
structures. Metallurgical and Materials Transaction A, 1999, 30A: 3153-3165.
[53] Dierk Raabe, Franz Roters, Frdric Barlat, Long-Qing Chen, Continuum Scale
Simulation of Engineering Materials: Fundamentals-Mcrostructures-Process
Applications, Wiley-VCH, Weinheim, 2004.
[54] B. Radhakrishnan and T. Zacharia. Simulation of curvature-driven grain growth by
using a modified monte carlo algorithm. Metallurgical and Materials Transactions A,
1996, 26(1): 167-180.
[55] S. Sista, Z. Yang, and T. Debroy, Three-dimensional Monte Carlo simulation of grain
growth in the heat-affected zone of a 2.25Cr-1Mo steel weld, Metallurgical and
Materials Transactions B, 2000, 31: 529-536.
[56] S. Mishra, T. Debroy, Measurements and Monte Carlo simulation of grain growth in the
heat-affected zone of Ti-6Al-4V welds, Acta Materialia, 2004, 52: 1183-1192.

Welding Processes 190
[57] Y.H. Wei, Y.L. Xu, Z.B. Dong, and J.L. Xia, Three dimensional Monte Carlo simulation
of grain growth in HAZ of stainless steel SUS316, Key Engineering Materials, 2007, 353-
358: 1923-1926.
[58] M.-Y.Li, E. Kannatey-Asibu, Monte Carlo simulation of heat-affected zone
microstructure in laser-beam welded nickel sheet, Welding Journal, March 2002: 37-44.
[59] Weibin Wang, Yaowu Shi, Yongping Lei, and Zhiling Tian, FEM simulation on
microstructure of DC flash butt welding for an ultra-fine grain steel, Journal of
Materials Processing Technology, 2005,161: 497-503.
[60] J. H. Gao, R. G. Thompson, Real time-temperature models for Monte Carlo simulations
of normal grain growth, Acta Materialia, 1996, 44: 4565-4575.
[61] http://www.ssab.com/Global/SSAB/SSAB_Americas/A514%20Grade%20S%20Brochure
%2006-25-08.pdf
[62] http://www.midalloy.com/html/pdf/low_alloy/TB%20ER100S-G%20Weld.pdf
Chapter 9




2012 Khairuddin et al., licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Principles and Thermo-Mechanical
Model of Friction Stir Welding
Jauhari Tahir Khairuddin, Jamaluddin Abdullah,
Zuhailawati Hussain and Indra Putra Almanar
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/50156
1. Introduction
Friction stir welding (FSW) is a solid-state welding process that gained much attention in
research areas as well as manufacturing industry since its introduction in 1991 [1, 2]. For
almost 20 years, FSW has been used in high technology applications such as aerospace to
automotive till high precision application such as micro welding. The main feature of a
solid-state welding process is the non-melting of the work material which allows a lower
temperature and a lower heat input welding process relative to the melting point of
materials being joined. This is advantageous over the conventional fusion welding where
excessive high heat input is required to melt the work material. Much less heat input
required for FSW translates into economic benefits, safer and less complicated welding
procedures. The friction stir welding make it possible to join light weight materials such as
aluminium alloy, magnesium alloy, copper and titanium alloys which are very difficult to
weld by conventional welding. These clear advantages have greatly increased the usage of
these materials in structural applications [3, 4]. In addition, FSW also makes possible to
produce sound weldment in 5000 and 7000 series aluminium alloys that are not possible to
be welded using conventional method. FSW does not produce sparks or flames. Thus,
safety, environmental and legislation issues are not of major concern. FSW process provides
proven good quality and strong weldment with inexpensive and lesser number of
equipment, eliminates the use of filler metal and improved weldability. Due to these factors
FSW has successfully been employed in aerospace, automobile and ship building industries.
The need to further understand and improve FSW process continues to propagate in many
applications. Many researchers have looked into several methods including mathematical
modelling of the process, aiming at understanding the physical-material interaction [8 14].
However, there is lack of recorded work in the literature on a system or method to
quantitatively measure the welding parameter such as force and torque in FSW process.

Welding Processes 192
This chapter aims to introduce FSW process, its parameters, the applications and a thermo-
mechanical model of the process. From the mathematical model derived, a measurement
apparatus was developed for force and torque determination during FSW process and
experiments were performed to validate the model.
2. Friction stir welding process
FSW set up consists of (1) cylindrical rotational tool, (2) two or more work materials of
similar or dissimilar material combinations (3) backing plate and finally (4) clamping or
holding fixture as shown in figure 1.
The rotating tool design consists of a combination of two cylinders of a specific radius ratio
known as shoulder and smaller radius pin or probe, where the height of the pin or probe is
usually more than half of the work material thickness but not equal to its overall thickness.
The work materials to be joined may be arranged as conventional welding method but the
most common configurations used in FSW are abutted and lapped configuration. For any
configuration, FSW has the capability to join thick plate without the need for special
preparation prior to welding process. Meanwhile, the backing plate is to ensure the
establishment of confine volume and it becomes a must when welding with a pin
penetration approaching the bottom of the work materials. The most crucial part of the work
materials set up is the clamping or holding fixture. Improper clamping may jeopardize the
mated surfaces to be welded and will generate gaps leading to the formation of worm hole
or cavities in the weldment. FSW is the non-filler process; hence no substitute material to fill
in gaps created by the separation of the work materials is required.

Figure 1. Friction stir welding experimental set up [5].

Principles and Thermo-Mechanical Model of Friction Stir Welding 193
2.1. Stages in friction stir welding process
FSW process involves four phases which are (1) plunging phase, (2) dwelling phase, (3)
welding phase, and finally (4) exit or retract phase. Briefly, the process starts with rotating
tool pin or probe thrusting onto the configured work materials under a constant axial load
to generate friction heat. This process will continuously increase the temperature at the
immediate contacting surface of the rotating tool and work material. The process continue
until the temperature at the immediate contact of the rotating tool and the work material
increased to a temperature which causes the work material to soften, plasticized and
significantly lose its strength. Consequently, these conditions allow the rotating tool to
penetrate to a certain depth usually almost to the thickness of work material. The plasticized
material is subjected to displacement by the rotating tool pin plunge, effectively being
flashed out with a portion of the generated heat, thus introducing new immediate lower
temperature and harder surface of work material. This event explains the transient heat
generated through pure mechanical friction work at the tool and work material interface.
The end of the plunging phase is signified by the sound contact of the rotating tool shoulder
with the immediate work material surface.
At this moment, the process enter the dwelling phase where the rotating tool is allowed to
dwell for a period of time, causing the temperature to increase further, up to its hot working
temperature. The heat generated from frictional work is greatly dependent on the relative
increase of contact surface area as well as the relative speed. The heat generated causes the
affected area under the shoulder to expand considerably. Phenomenally, the heat causes the
work material close to the immediate contact to lose its strength and becoming plastic. Once
this condition is reached, thin soft material layers is produced and would stick to the
dynamic rotating tool surfaces (pin and shoulder) and being forced to be displaced along.
Instantly the mechanical friction heat generation is partially turned into plastic dissipation
heat generation. It is explained by the energy dissipated from the internal shearing of
different velocity between the displaced soften work material layers to static more solid
surface. Ideally, intermittent heat generation mechanisms due to friction work and plastic
dissipation take place because of the transient heat transfer effect and the material ability to
regain its strength as heat is lost to the ambient. In addition, the other role of these frictional
work and plastic deformation mechanisms are to induce soft material displacement and
causes the stirring action or severe material deformation which later produce the
amalgamated joint.
The dwelling phase is followed by welding phase. After the local temperature of work
material under the rotating tool approaches its hot working temperature and is soft enough
to be stirred and displaced, the rotating tool is moved transversely along the welding line.
This traverse motion caused the plasticized soft material at the leading edge of the rotating
tool being squeezed and sheared through a small slit formed by the displaced soft material
at the side or lateral of the tool, preferably in the direction of tool rotation. The displaced soft
material is then deposited to the gap at the trailing edge left by rotating tool pin or probe.
The soft plasticized material is forcedly displaced by the rotating tool along its rotating

Welding Processes 194
direction under a closed encapsulation of harder solid work material wall and rotating tool
shoulder. The soft material is forged to the trailing edge in layers, forming weld nugget. At
each traverse increment of the rotating tool motion, the displacement of soft plasticized
work material to the trailing edge will introduce new solid, lower temperature work
material at the leading edge. Thus it reintroduces friction work heat generation mechanism
prior to plastic deformation mechanism and continuously repeating the heat generation
process all over again at each traverse displacement of the tool. This produces cyclic
transient heat generation. This cyclic process takes part throughout the welding phase and
strongly affected by the combination of the rotating tools rotational and traverse speed.
Recap, during the welding phase the plasticized material is subjected to displacement,
extrusion and shearing mechanisms facilitated by the tool rotation, thrust and transverse
movement under cyclic heat generation along welding line and finally consolidating
welding nugget in the trailing side.

Figure 2. Friction Stir welding process phases of a butted work material configuration [6].
At the end or exit phase of FSW process, the rotating tool is retracted away from the work
material leaving a cylindrical hole mark that once occupied by the tool pin. For cosmetic
reason, the cylindrical hole may be filled with filler material at the end of the welding
process but the most common method used is by introducing dummy material prior the exit
phase. Dummy material is of the same material used for the work material to be weld and
placed at the end of welding line where the rotating tool is allowed to traverse to and exit
within it. The dummy material is later to be cut away leaving good surface finish. Though,
this cosmetic issue would remain in the application of innovated friction stir spot welding.
These process phases in FSW are dependent to one another to produce a good amalgamated
weldment and are strongly affected by the welding parameters. The assurance of good
weldment is determined by proper control of varying measureable welding parameters such
as rotational speed, axial plunge force and torque, traverse speed, tool geometry and
Tool
Shoulder
Pin
Work material
Advancing side
Retreating side
Weldment
Pin hole

Principles and Thermo-Mechanical Model of Friction Stir Welding 195
orientation in the form of heat energy. Similar to other conventional welding methods, heat
energy notably determines the quality of the joint.
2.2. Thermo-mechanical dynamic of friction stir welding
To understand the working principle of friction stir welding, it is necessary to explore the
physics related to heat generation. Friction stir welding process started with initial
mechanical friction between tool and the surface to be welded which results in heat
generation. Rotating tool rotates and advances in trust motion against the work material
followed by heat generated through plastic dissipation or deformation of sheared and
strained layers of soften material close to the rotating tool surfaces. The heat generation
mechanisms occur intermittently or in cycle throughout the welding process.
2.2.1. Friction heat generation
Mechanical friction work is initiated when rotating tool surfaces are in contact with the
immediate stationary surface of work material under a normal load. While the rotating tool
is sliding, it introduces velocity difference between the dynamic rotating tool and the static
work material surface and thus creates friction work and subsequently, heat. This
mechanical friction work is described based on the Amontons laws which firstly explain
that the friction between two separate bodies is directly proportional to the normal load
applied onto the bodies. In this law, coefficient of friction of static friction is a constant
variable and temperature dependent but only to be considered as kinetic friction when the
contact condition is non-sticking or sliding. Secondly, the friction force is independent on
the apparent contact area [7].
In general application, friction work is assumed based on Coulombs dry friction model
between solid bodies, which is also at the same time conforms the aforementioned
Amontons law. In details, the mechanical work and heat generation relationship in the
presence of sliding friction is explained by contact conditions between hard and soft metallic
material interaction. It involves very small scale asperities at the contact surfaces of hard
and soft material. As normal force is acting on the rotating tool, it is being distributed onto
smaller area asperities at the contact surfaces which resulted in a very high pressure per unit
force. Due to relative velocity difference, the normal force causes ploughing of the soft
material by the hard materials asperities. The soft material gets agitated, deformed and
finally broken, releasing the stored energy in the form of heat [8].
The released energy which is a very high local thermal energy causes the temperature to
rise. The heat energy is eventually transferred and stored into the rotating tool and the lump
work material. The heat causes the work material to soften, reduces its strength, broken and
deforms into soft material layer in between the rotating tool and the work material. This soft
layer is gradually displaced by the rotating tool pin revealing new surface contact condition
for another cycle of ploughing action by the rotating tool. The process promotes another
cycle of mechanical friction heat generation.

Welding Processes 196
2.2.2 Plastic dissipation heat generation
This phenomenon typically occurs at higher temperature resulting from friction heat
mechanism but only significant during dwelling and actual welding phase where work
material is confined under the rotating tool. Through friction heat mechanism, work
material temperature under the rotating tool is increased to a degree where work material
layer at rotating tool interface started to lose it strength, yielded, stick and move with the
rotating tool. This phenomenon increases the thermal softening effect from friction heat and
causes shears at the work material to work material layers interface. It reduces friction heat
mechanism but at the same time introduce high strain rate plastic deformation. As a result,
highly localized heat is generated internally within the work material itself away from the
rotating tool to work material interface due to dynamic velocity differences and boundary
sliding condition [9, 10, 11].
Due to the heat transfer within the process, the plastically deformed material tend to recover
its strength thus establishing new lower temperature level and reinitiate mechanical friction
heat generation mechanism [12]. Furthermore at the welding phase, heat generation from
plastic dissipation mechanism is increased at cost of the travelling rotating tool and the
shearing of the work material to higher extend toward trailing edge. It further increases the
process temperature as the rotating tool is reintroduced to new contact condition at the
leading edge and reinitiates the mechanical friction heat generation mechanism [13].
These heat generation mechanisms throughout FSW process occur in cycles due to the
instance of slip and stick contact conditions and the alternating boundary conditions at the
material-rotating tool interface [14]. In this regards, heat generation mechanism has most
influential effects on the process introducing high strain rate and thermal effect which soften
work material adjacent to the tool to work material interface, to be encompassed the at the
trailing edge and produces welding nugget which posture the main characteristic of the
joint [15].
2.2.3. Heat transfer
Simultaneously, heat generated is constantly being transferred within the system; portions
of heat are distributed within the work material, the rotating tool, the holding fixture or
backing plate and finally to the ambient. In depth, heat generated is subjected to three
dimensional (3-D) heat flows away from the heat source under boundary conditions. Heat
input and heat transfer at the rotating tool are indirectly coupled at the rotating tool to work
material interface into the work material through heat conduction and incorporated with
convective heat transfer effect around the pin in the deformation zone [12, 13, 16, 17]. For the
work material, convective and radiative heat transfers are considered for heat exchange at
the top work material surface, past the shoulder peripheral while at the same time only
convective condition is considered for the bounding surfaces of work material [18, 19]. For
the case of backing plate, appropriate variable gap conductance is considered for work
material to backing plate interface depending on specific thermal contact resistance
condition; of temperature dependent or contact pressure dependent or surface contour of
work material or any of the combinations [13, 18, 20].

Principles and Thermo-Mechanical Model of Friction Stir Welding 197
Ultimately 100% energy generated from mechanical work throughout the process is
converted to heat and physical deformation with approximately 88% of heat is conducted
and distributed globally through to the lump work material, backing plate and else to the
rotating tool [12]. Heat plays a very significant role not only toward the physical success of
the joint but also towards the process temperature profile, heat transfer and the work
material internal strain and stress distribution. It causes direct influences on the final
weldment microstructure and residual stress resulting properties which are strongly
correlated to the welding variables and temperature-dependent material properties [21].
2.2.4. Friction stir welding mechanism
Weldment is produced in the course of welding phase where layers of materials are forced
to move along with and around the rotating tool surfaces at a specific contact condition;
fully sliding or sliding and sticking or fully sticking. The soft material layers motion are
heavily deformed and driven by the tool rotational direction. It is forced through the
retreating side toward trailing edge and downward closed to the pin before finally being
forged and deposited at the once occupied volume of the rotating tool pin at the trailing
edge under severe plastic deformation [17, 21, 22, 23].
The mechanical torque of the rotating tool causes mechanical shearing to the immediate
work material close to the rotating tool and work material interface, forcing soft work
material layers to motion and strained, creating flows of fine layers of materials prior
progression into weldment. The flow motion or velocity is visually estimated through the
grain size and shape, correlated to internal strain rate [24]. The material flow and joining
mechanisms are described by the region formed by the FSW process known as the flow
zones. It describes the zones where shear layers are visibly distinguished by material
characteristics and exhibit the evident of non-melting but severely deformed soft material
deposited into amalgamated weldment in layers and flowing manner.
2.2.5. Welding characteristics
Temperature profile and history of FSW process are resembled by the distinct regions at the
weldment. These regions are characterized by discrete microstructure sizes, shapes and
varying properties, produced by the significant thermal effect and mechanical deformation.
Under the heat generation, lump effect and heat transfer of the process, thermal profiles are
being distributed from the crown shaped heat source around the rotating tool to work
material interface toward the peripheral work materials surfaces and edges [22, 25]. These
regions are known as; 1) weld nugget, the product of plastic deformation due to the stirring
effect deposited behind the rotating tool pin at the trailing edge, 2) thermo-Mechanically
Affected Zone (TMAZ) of internally sheared plastic deformation within the work material
away from rotating tool to work material interface, 3) heat Affected Zone (HAZ) of
structurally altered and thermally affected region due to intense temperature different
between TMAZ and base metal temperature region, and 4) base metal of work material
which is not physically affected by the thermal effect [26].

Welding Processes 198
Temperature profile within the FSW process portrayed direct relationship of the heat
generation, torque generated through the rotating tool, loads exerted throughout the work
material and power consumed by the welding process. Both thermal and mechanical effect
from heat generation and stirring effect engender welding characteristic in term of stresses,
tensile and hardness properties. For which, the rotating tool rotational and traverse speed
interchangeably influenced the temperature profile and affectively manipulate the material
flow behaviour, weld material chemical composition, microstructure orientation, strain,
residual stress, thermal stress, material hardness and strength of the weldment [27, 28, 29]
2.3. Applications of friction stir welding
Recently, FSW won attractive attention by manufacturing industries because of the ability
that outperformed other welding technique such as tungsten inert gas (TIG) to weld
aluminium. Besides, it has the ability to be adapted in advanced automation system such as
robotic which is very predominant in aluminium components and panel fabrications of
highly rated technology readiness level [30]. For example, it is an applicable technique
adapted for rail cars for fabrication of floor panel part of a Type 700 Shinkansen or bullet
train, as well as the aluminium roof, side wall and floor panel for suburban train and other
more recent commuter or express rail cars [31]. This is due to its low distortion and its
suitability to produce large welded products including prefabricated panels as well as
tailored blanks and joints. These advantages are also shared in marine application when it is
first commercially used for ship building in 1996 with the ability to joint thick panels,
sandwiched, honeycomb panels and corrosion resistance material panels [32]. FSW is well
favoured for performing butt join in comparison to conventional arc welding application
which also turns out to be significantly viable in terms of low labour cost and shorter
welding cycle time [33].
The other important application of friction stir welding is in aeronautical and aerospace
industries where aluminium alloy is used as primary material for their construction. The
welding process enables manufacturers to completely replace the riveted joints and
assemblies of lapped and abutted configuration that are used mainly for fuselage sections,
propellant and fuel tanks of commercial air craft as well as space launch vehicle [34]. Thus,
the process allows total elimination of the use of thousands of rivets. This has resulted in
better quality, stronger and lighter joints at reduced assembly cost for aviation industry.
Meanwhile in automotive application, FSW and its recently innovated process of friction stir
spot welding (FSSW) are introduced to replace conventional resistive spot welding (RSW) in
the transition when aluminium alloy application is ready and becoming ideal to be used for
panels in passenger and commercial vehicle bodies. Aluminium application in automotive
industry are sought for its prefabricated and tailored panel, good strength-to-weight ratio,
potential for reducing fuel consumption, its ease of recycling as well as marked reduction in
production cost. This has compelled car manufacturers to use the same concept not only for
the body panels, but for other parts as well [35].

Principles and Thermo-Mechanical Model of Friction Stir Welding 199

Figure 3. Tailored welded blanks in a passenger vehicle [36].





Figure 4. A component of friction stir tailored welded blank [36].




Since its introduction, friction stir welding has matured significantly with widespread use of
wide range aluminium alloys for structural applications with its compatibility to be used in
ferrous, stainless steel, nickel, copper and titanium alloys. However, only a small percentage
of world welding and joining market has implemented the process. It is still relatively
underutilised and this is varied among industries, university researches, collaborations and
other niche applications. With all the challenges to joints wide variety hard to weld
materials, the motivation of the application and adaptation of this noble technology in
manufacturing not only to meet quality joints, but for the potential economic value and also
for its environmental friendliness.
2.4. Recent development & future outlook
Since the last two decades, there are considerably vast lab and industrial work done on FSW
process which leads to the emerging advancement of new materials application and
combinations, process improvement, tool designs, welding configurations, tailored blank
application and adaptation to automation. Though, the most important innovation in the
process itself is its variations; (1) high speed FSW (HS-FSW), (2) ultrasonic stir welding
(USW), (3) thermal stir welding (TSW), (4) friction stir spot welding (FSSW), (5) friction stir
joining (FSJ), (6) friction stir processing (FSP), (7) friction bonding (FB) [37].

Welding Processes 200
The recent advancement shows the introduction of HS-FSW aiming at reducing the process
forces by means of increasing heat generation rate. This reduction in forces is to realize the
idea of permitting manual handheld welding work. HS-FSW process is also designed to
achieve lighter and portable device or equipment. This will make possible manual handheld
welding and eliminates the need for rigid fixturing of the work material and rotating tool.
USW is another variation of the process where ultrasonic energy is used to assist in initial
heat generation. The objectives are to reduce process forces and welding time. The major
boost of USW is it reduces dependency on the tool shoulder to generate heat and instead,
make use of the coupled ultrasonic vibration to further agitate the rotating tool at the contact
interface, amplifying the mechanical friction effect.
Different from USW that utilize ultrasonic energy to generate heat, TSW decouples heat
generation from mechanical friction of conventional FSW and instead, utilizes external
induction heating to increase work material temperature at the welding spot or welding
line. Though induction heating may ultimately increase the rate of heat generation but it
reduces strain rate of plastic deformation, a prevalent characteristic in FSW. Reduced strain
rate will significantly affect the final welding properties [38].
The most common variation of all is the FSSW, which is well known to be used in
transportation industries to replace rivet and as well as adhesive joining method. Though
this method produces cosmetic defect in the form of pin hole created by the rotating tool, the
weldment has sufficient mechanical strength for joining method [39].
While the other FSW variants are dedicated for metallic materials, FSJ method is dedicated
for joining thermoplastic materials. It has been used for polypropylene, polycarbonate and
high density polyethylene materials [40. The advancement of FSJ method may possibly
makes way for mechanical joining method involving plastic matrix composites.
FSP is a unique non-joining method variant of FSW where it utilizes the stirring process to
alter microstructure derived from FSW method to be super fine, modified with improved
physical properties and at the same time suppressed defects such as porosity and short
crack [34]. The introduction of foreign particle into base material during FSP creates new
near quality to metal matrix composite or MMC structure and thus improving the material
properties. This provides a new platform to produce improved future materials.
The consequent of FSP method emerges as a modified method of FB that allows bonding of
overlapping thin plates through locally modified microstructure. The process utilizes less
stirring effect due to the short pin geometry and thus differs from common FSW method.
The majority of these FSW variants stress on their ability to increase welding time and at the
same time to reduce forces exerted from the welding process. The other purpose of these
methods is to provide means of producing joining method involving different type of
materials and configurations. More or less, these method share the same mechanism of
mechanically manipulating soften work material and formation of fused material at the
trailing edge.

Principles and Thermo-Mechanical Model of Friction Stir Welding 201
3. Friction stir welding parameters
Independent process variables play significant effect on the welding process and the process
control. The process variables entail the axial force for plunging, rotating tool rotational
speed, rake angle, welding speed and tool geometry. The aforementioned variables strongly
affect the heat generation rate, temperature profile within the work material, mechanical
power required by the process, material evolution of the weldment and also loads
distributed within the work material. These variables are extensively discussed in this
chapter to understand the mechanics of joining, process and final weld properties
optimizations, where direct measurement can be done experimentally and predicted
numerically.
3.1. Effect of welding parameters on joint quality
It has been reported that rotational and traverse speeds have both direct and indirect
influence to the final weldment. Its direct influences to the mechanics of joining suggesting
the degree of stirring based on the contact condition and multi component loads. Whilst, its
indirect influences to mechanical properties of the weldment is heavily derived from the
combination of temperature exposure and tool design [41]. Low rotational speed induces
low stirring effect and finally low heat generation rate. At low traverse speed, it increases
exposure to heat source, facilitate more material flow and reduced multi component loads,
vice versa to high traverse speed but only to an extent. Extreme high rotational speed results
too much heat while extreme high traverse speed results less heat, low stirring effect and
increase travel resistance due to heavy loads on the rotating tool and hard work material at
the leading edge. At this moment, the material is sheared to the lateral side instead of
moving around the rotating tool direction [42]. Excessive heat which is generated from
either high rotational speed or low traverse speed, or combination of both conditions,
significantly reduces the mechanical properties of joint due to microstructure evolution of
the regions exposed to excessive heat [43]. Albeit, the appropriate traverse and rotational
speed below critical speeds might result optimum heat generation rate and reduced thermal
exposure that produce good strength and hardness of weld joints [44, 45].
Torque produced during the welding process depends heavily on the contact conditions
which are determined by the rotating tool rotational speed, degree of softness of work
material or plasticity at the rotating tool to work material interface, axial load exerted and
the tool design [22, 46, 47]. Any changes in the traverse speed at a constant tool rotational
speed do not significantly affect the temperature profile compared to changes done in the
rotational speed. These make the torque to be insensitive to the traverse speed. In addition,
optimum rotating tool design influences torque produced through the effect of sum of
contact areas and contact conditions at the interface where it plays major role in plastic
deformation or strain work distribution, material transportation or flow and toward process
loads and work material temperature [41, 48, 49, 50]. As temperature increases, work
material temperature dependent shear stress plump and no longer display full solid
properties thus reduces the torque at the rotating tool to work material interface and further
reducing power and energy required to produce heat within the process.

Welding Processes 202
3.2. Optimizing process parameters
Considerably vast works has been made in pursue to understand and optimize the physical
process of FSW that influenced by the associated variables using both empirical as well as
numerical models for the heat generation, material interaction and flow.
3.2.1. Thermal modelling
Heat generation is modelled based on the torque required to rotate a circular shaft relative
to the material surface. The model is made by assuming a constant coefficient of friction,
pressure distribution and 100% conversion of the shearing work to heat. It is also assumed
that the net power required is directly proportional to the tool rotational speed and the tool
shoulder radius i.e. q R [51].
Friction work principle used for heat generation model is coupled with plastic work
principle to model three-dimensional heat and material flow based on temperature
dependent coefficient of friction and temperature dependent pressure distribution for
aluminium alloys [26]. In detail, three-dimensional visco-plastic flow and heat transfer have
been investigated through solving the equations of conservation of mass, momentum and
energy by considering the heat source at two separate conditions. Firstly, visco-plastic flow
and heat transfer at the tool to work material interface due to the mechanical friction or due
to the plastic dissipation heat generation mechanisms and secondly, the visco-plastic flow
and the heat transfer within the layers of soft work material away from the tool to work
material interface under the combination of both friction and plastic dissipation heat
generation mechanisms [11, 52, 53, 54].
The model is defined by the contact area, the radial distance of the rotating tool pin and
shoulder, the material shear stress or the spatially variable coefficient of friction or their
combinations, the angular velocity and the exerted normal pressure acted on the work
material surface. Then, the model is validated through comparison of the computed heat
generation, peak temperature and the total torque exerted on the tool with the experimental
results. Contact conditions at the interface are described as sliding, sticking or partial
sliding or sticking, known as slip factor is also being adapted. The slip factor is derived
experimentally, determined by the plunge force and torque from the welding process. The
slip factor yield a proportional relationship between plunge force and heat generation
where Coulombs law of friction is applied to describe the shear forces at the interface [10,
55]. Slip factor is also used to evaluate the welding energy and temperature, utilizing
torque based heat input [53]. These works allow the prediction of the welding
temperature from the transverse speed, tool rotational speed and the applied force.
Ultimately, fully coupled thermo-mechanical model with adaptive boundary condition
which applied both thermal and mechanical model is used to predict transient
temperature profile, active developed stresses as well as the three-dimensional force
components [13].

Principles and Thermo-Mechanical Model of Friction Stir Welding 203
3.2.2. Model validation
The force and temperature measurement experiments are conducted under different
welding parameters for model verification by differing the transverse and the tool rotational
speed whilst maintaining the constant vertical force. The result is later used for the
calculation of the heat input into the tool and workpiece. In relation, numerical model is
developed to take account the process effects on the work material such as the material
plasticity properties, thermal expansion and stress, cooling effect, stress stiffening, stress
distribution, material strain, residual stress as well as the thermal history o the welding
process [56, 57, 58].
The mechanical effect is visualised and simulate as material flow model to describe the
process parametric effects on the soft work material flow and the welding mechanisms [59,
60]. The physical material flow highlights the particular material strain, its distribution and
flow motion around the rotating tool, at the leading edge of advancing and retreating sides
and at the trailing edge as well as the actual bond that might occur in the FSW process.
Thus, thermo-mechanical model is required to exhibits the importance of the three-
dimensional loads and torque exerted by the rotating tool material to determine the best and
optimized parameter for the welding process of any materials which is greatly related to the
joint and process properties such as residual stresses, process temperature, joint strength
and productivity. The parameter optimization is related to variables such as welding speed,
rotating tool rotational speed as well as rotating tool design. Proceeded, welding power has
been modelled to determine the overall heat input for FSW process based on the traverse
speed and tool rotational speed as well as its effect on the material properties. The model
has been derived based on the relationship between the rotating tool rotational speed and
the represented rotating tool torque with association to the key parameters aforementioned
[61, 62].
Significantly all of these works and models represent the correlation of the mentioned
independent variables of the associated welding parameters as well as material properties
serve the heat generation, temperature distribution and joining mechanism that contribute
to the success of the welding process. In summary, the relationship between the
independent process variables and the dependent process output to the heat generation
mechanisms is best described as in a closed and correlated system, explained by the
interaction of process welding variables and key process conditions; physical, metallurgical,
heat generation and heat transfer effect [63].
4. Mathematical modelling of welding forces & torque
FSW is not widely available in general application rather than conventional fusion joining
methods especially in automotive industry due to shorter process and heat generation cycle.
Thus, the true understanding of FSW process is still in far-reaching; understanding the
physics and nature of the process, lacking of standards guideline and practice, optimization
of the process for typical material usage or application as well as still typically not suitable
for small scale robotic or manual handling. The mathematical model on the welding forces

Welding Processes 204
and torque is proposed for optimizing FSW process and to study the effects of varying its
parameters. The model is to be compared with experimental work [64, 65].
4.1. Model development & assumption
Initial heat generation takes place at the first contact of the rotational tool pin surface and
continue throughout the plunging phase where the temperature distribution of the work
material is asymmetrical at the leading, trailing edge as well as advancing and retreating
side. It is based on the assumption that the interface heat generation is constant with the
consideration of the constant rotating tool angular speed, , temperature-dependent
pressure distribution function, P(T), heat capacity, cp, thermal conductivity, k and constant
coefficient of friction, k. Based on Fouriers 2
nd
law [66];

2 2 2
2 2 2
p
T T T T
c k q
t
x y z

| |
c c c c
= + + + |
|
c
c c c
\ .
(1)
Where cp is the heat capacity, x, y, and z are the space coordinate and is the heat source
term correspond to heat generated from the welding process. The function of the heat
generation is directly related to friction work of the contacting surface thus accounts the
sum of contact surface area of the tool. The sum or contact area is represented as function of
rotating too plunge depth, h. The data used is as in table 1. Given;

2
( 0) h p
A r t
=
= (2)

(0 5) ( 0) (0 5) 2
sin2
p p p
h h p p
p
R R r
A A h r r s
h
t
< < =
( | || |
( | | = + +
| |
(
\ . \ .
(3)

2
( 5) (0 5)
( )
h h s p
A A R R t
= < <
= + (4)
For A is the contacting surface between the rotating tool and the work material surface, rp is
the bottom pin radius, Rp is the top pin radius, hp is the total height of the pin and rpS2 is
the minor cone area.
Furthermore, the torque required to rotate the rotational tool relative to the static workpiece
surface under P(T) represents the conversion of mechanical work of the rotating tool. Given;

( )
2
0 0
( )2 ( ) ( )
R
M R
M dM P T A h rdr P T A h R = = =
} }
(5)
Or by simplification of the sum of contact area through cylindrical approximation, given [67,
68];

2 3
0 0
2
( )2
3
R
M R
M dM P T r dr PR t t = = =
} }
(6)

Principles and Thermo-Mechanical Model of Friction Stir Welding 205
Where M is the interfacial torque of the in contact workpiece surface and rotating tool
surface, k is the coefficient of friction, R is the contact surface radius, and P(T) is
temperature dependent pressure distribution across the interface.
For fully sliding contact condition and with assumption of all the friction work is converted
into frictional heat, the average heat input per unit area and time becomes;

2 3 2 3
0
0 0
2 4
( )2
3 3
R
M R
q dM P T r dr PR nPR

(7)
Where q0 is net power in Watt (Nms
-1
) and is rotational speed (rads
-1
). Apparently, in
equation 7, heat input depends on the normal pressure distribution function, contact surface
radius, temperature dependent function of coefficient of friction and the rotational speed of
the rotating tool which produce transient heat generation, distributed into the work material
and thus characterized the process variables for the friction stir welding process.
In order to understand the physics of the welding process in the expression of the
mechanical loading associated to the welding process, a mathematical representation is
derived based on figure 5;

1 1 1 1
( ) r x x i y j z k (8)

2 2 2 2
( ) r x x i y j z k (9)

3 3 3 3
( ) r x x i y j z k (10)

4 4 4 4
( ) r x x i y j z k
(11)

Figure 5. Free body diagram of multi-component load measuring device [65].

Welding Processes 206
0 F =

(12)

=
=

4
*
, , , ,
1
x y z x y z
n
F Fn (13)

* * * * * *
1 2 3 4 1 2 3 4 1 2
* * * *
3 4 1 1 2 2 3 3 4 4
sin cos ( ) (
) ( )
x x x x x x x x y y
y y z z z z z z z z
F i F k F F F F F F F F i F F
F F j F F F F F F F F k
u u + = + + + + + + +
+ + + + + +
(14)
Where F*, is the plunge force as the function of contact surface area and under the
temperature dependent pressure distribution P(T) for a desired depth of penetration. The
torque exerted on the work material by the rotating tool mechanical friction work is
calculated based on equation 6 and coupled by the moments reacted at each of the
measuring references on the workpiece as in figure 5;
0 M =

(15)

4
*
, , , , , , , , , ,
1
4( )
x y z x y z x y z x y z x y z
n
M rn Fn Rn Fn
=
= =

(16)
=
* *
, , , ,
2
3
x y z x y z
M F R (17)
Where;
= = =
1 2 3 4
M M M M (18)
= = =
1 1 2 2 3 3 4 4
r F r F r F r F (19)
In figure 5, the moment exerted on the measuring references is in equilibrium and the acting
forces exerted by the rotational tool mechanism are determined from equation 18, generally;
= =
*
, , , , , ,
1
; 1, 2.., 4
4
x y z x y z x y z
M Rn Fn n (20)
= + + ( ) ( ) ( )
n n n z n y n x n y n x
r F y Fn z Fn i x Fn j x Fn y Fn k (21)
The general equation can be presented in a matrix form as;

(
(
( (
(
( (
=
(
( (
(
( (


(

*
*
1
0
4
0 0
1 0
4
x
n n x
n n y
n n z
z
M
z y Fn
z x Fn
y x Fn
M
(22)
At any static equilibrium where the slip contact condition between the rotating tool and the
workpiece surface remain constant, the moment at each of the measuring references remain

Principles and Thermo-Mechanical Model of Friction Stir Welding 207
the same but dependent on the radius of the rotating tool and the function of pressure
distribution P(T) for the respective rotating tool position. The values for the moment are also
remained constant at any Cartesian coordinate (x, y) location of the rotating tool on the
surface of the workpiece. Although the basic assumption for the constant contact condition
for the function of pressure distribution P(T) in modelling the process is not physically
correct but appropriate to be acceptable in the context of numerical model as average value
used throughout the investigation.

Properties/parameter
Work material dimension, mm
Shoulder radius, mm
Tool radius, mm
Pin radius, mm
Pin height, mm
Pin conical angle,
Tool angle,
Workpiece material
Tool material
Coefficient of friction
Plunge forces
Value
200 X 200 X 7
9
3
2.852
5
2
2
6061 T-6
M42
Figure 7
Figure 8
Table 1. Summary of data used for loads and torque calculation [5].
4.1.1. Co-planar analysis
A closer approximation is made on the geometry of the work material is to measure the
reaction forces and moment caused by the rotating tool from the friction welding process.
Equation 22 is reduced by performing co-planar analysis at measuring points of the workpiece
as in figure 6 for the initial heat generation until the full penetration of the stirrer tool.

Figure 6. Co-planar analysis of the workpiece material during welding process [65].

Welding Processes 208
Parametric analysis are carried out based on the work material temperature dependent
tensile, yield and shear strengths properties, welding configuration, work material
dimension and also tool design to investigate loads and torque distributed within the work
material. The temperature dependent material properties data used in the calculation are as
in figure 7 and 8 used to determine the theoretical plunge forces for the FSW process. In
addition, a constant plunge force is proposed as to simulate the experimental work as
comparison to numerical analysis.
Theoretically in this work, the mechanical properties suggest the controlling parameters of
the welding process especially regarding the plunge force which is based on the contact area
and controlled pressure parameter on the tool, heat energy consumed and the mechanical
loads applied.
The details of the parametric analysis are explained based on the case studies as carried as;
Case 1: Theoretical plunge force based on temperature dependent tensile strength. This
approach is based on the assumption that contact condition is fully sliding follows the
temperature dependent coefficient of friction curve as in figure 7.
Case 2: Theoretical plunge force based on temperature dependent yield strength. This
approach is also based on the assumption that contact condition is fully sliding.
Case 3: Theoretical plunge force based on temperature dependent shear strength. The
assumption for fully sticking contact condition is that the rotating tool which is trusting and
work material is separated by a thin layer of plasticised material at the contact interface.
Case 4: Experimental plunge force based on manual force and plunge depth control method
[5]. Force and plunge depth control methods are common control method for FSW process
[69, 70].

Figure 7. Temperature dependent coefficient of friction [5].

Principles and Thermo-Mechanical Model of Friction Stir Welding 209

Figure 8. Variable temperature dependent material properties of aluminium alloy 6061 and plunge
pressure variation for FSW process cases 1 4 [5, 6].
Though in this work, only case 1, 3 and 4 are being considered to visualize the maximum
and minimum limits of the parametric effect in comparison to the experimental work. Case 1
encompasses the maximum strength value of the work material prior failure due to the
mechanical effect of the FSW process. As for case 3, it represents the minimum requirement
of the FSW process to initiate mechanical or physical effect on the work material. Finally,
case 4 is the corresponding load suggested based on the work holding fixture as in figure 1
[5, 65].
4.2. Model validation with experimental data
The plunge variation profiles in FSW processes based on the temperature dependent work
material properties and experimental plunge pressure value are as exhibited in figure 8. It is
noticeable that the cases are significant at temperature range approximately 120C to 220C,
where the theoretical plunge pressure started to decrease. The work material becomes soft and
lost its strength, allowing the rotating tool to displace axially into the work material. The
corresponding rotating tool plunge force theoretical calculation based on the plunge pressures
schemes aforementioned are compared to experimental result and exhibited in figure 9.
During the plunging phase, the rotating tool thrusting under the plunge pressure profiles
and constant rotational speed produces mechanical torque. It acts on the work material and
initiates the mechanical friction work thus results the friction heat. The variations of
mechanical torque exerted at the rotating tool and the work material interface during the
plunging phase are as in figure 10.
Sticking and sliding contact conditions referred as case 1 and 3, exhibit significant torques
built up beyond 300C. For case 4, high torque exerted at initial plunge phase abruptly

Welding Processes 210
decrease as the temperature approaches 220C before it increase back at temperature beyond
300C. The decrease in torque indicate the softening effect of the work material while the
immediate increase of the torque values are due to the rotating tool shoulder surface comes
into contact to the work material.

Figure 9. FSW plunge forces as controlling parameters [5].

Figure 10. Tool torque exerted at the tool - work material interface for case 1, 3 & 4 [5].
An approximation of the process is based on case 4 where plunge pressure is based on
manual control of constant plunge force. The comparison of the theoretical plunge forces
and experimental work of the FSW process reveal distinct variation based on the
assumption used. However, the values act as references for the actual FSW process that
agree to the experimental result as shown in figure 9. The experimental result shows that its

Principles and Thermo-Mechanical Model of Friction Stir Welding 211
profile is approximate to case 4 plunge force scheme only at lower value. As for the rotating
tool torque, the experimental value remains closed to the value exerted by case 3 scheme at
almost constant trend. It described the nature of material in respond to the heat and
mechanical friction during the FSW process.
5. Conclusion
FSW process benefits solid state joining method that has great advantage on light weight
material such as aluminium alloy due to its thermal properties which make it difficult to be
joined using conventional methods. Similarly to the other welding method, heat generation
and heat transfer play major role in determining the success of the joining process as well as
predominantly establish the joint characteristics and properties. Though the detail of the
process mechanism and the effect on the welding has been widely studied in lab scale, good
understanding of the process mechanism provides a better view on choosing the best
parameter for the process and finally to achieve the best result in practice.
The future outlook of the process is very promising with new interest on its recent
development that allows broader application in term of material used as well as process
improvement. In addition, the development of mathematical analysis provides the ability to
predetermine the effect of parametric study of the process effect on the work material at a
shorter time as well as to be adapted to process automation.
Author details
Jauhari Tahir Khairuddin and Jamaluddin Abdullah
School of Mechanical Engineering, Universiti Sains Malaysia, Nibong Tebal, Penang, Malaysia
Zuhailawati Hussain
School of Materials and Mineral Resources Engineering, Universiti Sains Malaysia, Nibong Tebal,
Penang, Malaysia
Indra Putra Almanar
Mechanical Engineering Department, Universiti Teknologi Mara, Permatang Pauh, Penang,
Malaysia
Acknowledgement
This present work is supported by Universiti Sains Malaysia through RU-Grant (814084),
Universiti Sains Malaysia Institute of Postgraduate Studies Graduate Research Fund (IPS-
GRF) and USM-Fellowship scheme, which are greatly acknowledged.
6. References
[1] Thomas WM, Nicholas ED, Needham JC, Murch MG, Templesmith P, Dawes CJ.
Friction Stir Butt Welding, GB Patent No. 9125978.8; 1991.

Welding Processes 212
[2] Dawes CJ, Thomas WM. Friction Stir Welding of Aluminium Alloys. TWI Bulletin;
1995; 6 : 124-127.
[3] Hwang YM, Kang ZH., Chiou YC, Hsu HH. Experimental Study on Temperature
Distributions Within the Workpiece During Friction Stir Welding of Aluminium Alloy.
Int J of Mach Tools Manu; 2008; 48 : 778-787.
[4] Kim D, Badarinarayan H, Kim JH, Kim C, Okamoto K, Wagoner RH, Chung K.
Numerical Simulation of Friction Stir Butt Welding Process for AA5083-H18 Sheets. Eur
J Mechanics A-Solid; 2010; 29 (2) : 204 215.
[5] Jauhari TK. Development of Multi-Component Device for Load Measurement and
Temperature Profile for Friction Stir Welding Process [M.Sc Thesis]. Penang: Universiti
Sains Malaysia; Unpublished. 2012.
[6] American Welding Society. Welding Handbook, 9
th
Edition, Volume 3: Welding
Process; AWS; 2007.
[7] Popov VL. Contact Mechanics and Frictions: Physical Principal and Application.
London: Springer; 2010.
[8] Sherwood BA, Bernard WH. Work and Heat Transfer in the Presence of Sliding Friction.
Am J Phys. November 1984; 52 (11) : 1001-1007.
[9] Ravichandran G, Rosakis AJ, Hodowany J, Rosakis P. On the Conversion of Plastic
Work Into Heat During High-Strain-Rate Deformation. Shock Compression of
Condensed Matter Meeting. American Physical Society; June 24-29, 2001 . p 557 562.
[10] Schmidt H, Hattel J, Wert J. An Analytical Model for the Heat Generation in Friction
Stir Welding. Model Simul Mater Sc. 2004; 12 (1) : 143-157.
[11] Nandan R, Roy GG, DebRoy T. Numerical Simulation of Three-Dimensional Heat
Transfer and Plastic Flow During Friction Stir Welding. Metal Mater Trans A. April
2006; 34A : 1247 1259.
[12] Schmidt HB, Hattel JH. Thermal Modelling of Friction Stir Welding. Scripta Mater.
2008; 58 : 332 337.
[13] Soundararajan V, Zekovic S, Kovacevic R. Thermo-Mechanical Model With Adaptive
Boundary Conditions for Friction Stir Welding AL 6061. Int J Mach Tool Manu. 2005; 45
(14) : 1577-1587.
[14] Schneider J, Beshears R, Nunes Jr. AC. Interfacial Sticking and Slipping in the Friction
Stir Welding Process. Mat Sci Eng A. 2006; 435 436 : 297 304.
[15] Colligan KJ, editors. Friction Stir Welding: From Basic to Applications 2. The Friction
Stir Welding Process: An Overview. CRC Press; 2010. p 15 38.
[16] Khandkar MZH, Khan JA. Thermal Modelling of Overlap Friction Stir Welding for Al-
Alloys. J Mater Process Manuf Sci. 2001; 10 (2) : 91 105.
[17] Schmidt H, Hattel J. Modelling Heat Flow Around the Tool Probe in Friction Stir
Welding. Sci Technol Weld Joining. 2005; 10 (2) : 176 186.
[18] Khandkar MZH, Khan JA, Reynolds AP. In: Prediction of Temperature Distribution and
Thermal History During Friction Stir Welding: Input Torque Based Model. Sci Technol
Weld Joining. 2003; 8 (3) : 165 174.

Principles and Thermo-Mechanical Model of Friction Stir Welding 213
[19] Song M, Kovacevic R. Thermal Modelling of Friction Stir Welding in A Moving
Coordinate System and its Validation. Int J Mach Tools Manuf. 2003; 43 : 605 615.
[20] Shi QY, Dickerson T, Sherchliff HR., Thermo-mechanical Finite Element Modelling in
Friction Stir Welding of AL 2024 Including Tool Loads. In: Proceeding of the 4
th

International Symposium on Friction Stir Welding. 2003.
[21] Nandan R, DebRoy T, Bhadeshia HKDH. Recent Advances in Friction Stir Welding
Process, Weldment Structure and Properties. Prog Mater Sci. 2008; 53 : 980 1023.
[22] Arora A, Nandan R, Reynolds AP, DebRoy T. Torque, Power Requirement and Stir
Zone Geometry in Friction Stir Welding Through Modelling and Experiments. Scripta
Mater. 2009; 60 : 13 16.
[23] Arora A, Zhang Z, De A, DebRoy T. Strains and Strain Rates during Friction Stir
Welding. Scripta Mater. 2009; 61 : 863 866.
[24] Jata KV, Semiatin SL. Continuous Dynamic Recrystallization During Friction Stir
Welding of High Strength Aluminium Alloys. Scripta Mater. 2008; 43 (8) : 743 749.
[25] Song M, Kovacevic R. Numerical and Experimental Study of the Heat Transfer Process
in Friction Stir Welding. In: Proceedings of the Institution of Mechanical Engineers, Part
B: J Eng Manu. 2003; 217 (1) : 73 85.
[26] Dong P, Lu F, Hong JK, Cao Z. Coupled Thermomechanical Analysis of Friction Stir
Welding Process Using Simplified Models. Sci Technol Weld Joining. 2001; 6 (5) : 281-
287.
[27] Elangovan K, Balasubramanian V, Babu S. Developing an Empirical Relationship to
Predict Tensile Strength of Friction Stir Welded AA2219 Aluminium Alloy. J Mater Eng
Perform. 2008; 17 (6) : 820 830.
[28] Moreira PMGP, Santos T, Tavares SMO, Richter-Trummer V, Vilaca P, de Castro PMST.
Mechanical and Metallurgical Characterization of Friction Stir Welding Joints of
AA6061-T6 with AA6082-T6. Mater Design. 2009; 30 : 180 187.
[29] Tutum CC, Hattel JH. Optimisation of Process Parameters in Friction Stir Welding
Based on Residual Stress Analysis: A Feasibility Study. Sci Technol Weld Joining. 2010;
15 (5) : 369 377.
[30] Smith C, Hinrichs J, Crusan W. Robotic Friction Stir Welding: The State-of-the-Art. In:
Proceeding of the 4th International Friction Stir Welding Symposium. Salt Lake City,
UT; May 2003.
[31] Kumagai M, Tanaka N. Application of friction stir welding to welded construction of
Aluminum alloys. J Light Met Weld Constr. 2001; 39 : 22-28.
[32] Kallee SW, John D, Nicholas ED. Railway Manufacturers Implement Friction Stir
Welding. Weld J. 2002; 81 (10) : 47 50.
[33] Delany F, Kallee SW, Russell MJ. Friction Stir Welding of Aluminium Ships. In:
International Forum on Welding Technologies in the Shipping Industry. Shanghai,
China; June 2007.
[34] Arbegast WJ. Friction Stir Welding After a Decade of Development. Weld J. 2006; (3) :
28 35.

Welding Processes 214
[35] Kallee SW , Kell JM, Thomas WM, Wiesner CS. Development and Implementation of
Innovative Joining Process in The Automotive Industry. In: DVS Annual Welding
Conference. Essen, Germany; September 2005.
[36] ESAB. Friction Stir Welding the ESAB Way, XA00123720 [Internet]. 2004 [cited 24 May
2012]. Available from:
http://www.esab.co.kr/esab/mc_data/pdf/Friction%20Stir%20Welding.pdf
[37] Nunes AC. Friction Stir Welding. In: Schwart MM, Innovation in Materials
Manufacturing, Fabrication, and Environmental Safety, Boca Raton, FL . CRC Press.
2011: 88 119
[38] Ding J, Carter R, Lawless K, Nunes AC, Russell C, Suits M. Friction Stir Welding Flies
High at NASA. Weld J. 2006; 85 (3) : 54 60.
[39] Gerlich A, North T. Friction Stir Spot Welding of Aluminum Alloys. Canadian Weld
Assoc J, IIW Special Ed. 2006; 57 58.
[40] Mishra RS, Ma ZY. Friction Stir Welding and Processing. Mat Sci Eng. 2005; 50 (1-2) : 1-
78.
[41] Elangovan K, Balasubramanian V. Influences of Tool Pin Profile and Tool Shoulder
Diameter on the Formation of Friction Stir Processing Zone in AA6061 Aluminium
Alloy. Mater Design. 2008; 29 : 362 373.
[42] Zhang Z, Liu YL, Chen JT. Effect of Shoulder Size on the Temperature Rise and the
Material Deformation in Friction Stir Welding. Int J Adv Manuf Tech. 2009; 45 : 889
895.
[43] Liu HJ, Shen JJ, Huang YX, Kuang LY, Liu C, Li C. Effect of Rotation Rate on
Microstructure and Mechanical Properties of Friction Stir Welded Copper. Sci Technol
Weld Joining. 2009; 14 (6) : 577 583.
[44] Ren SR, Ma ZY, Chen LQ. Effect of Welding Parameters on Tensile Properties and
Fracture Behavior of Friction Stir Welded Al-Mg-Si Alloy. Scripta Mater. 2007; 56 : 69
72.
[45] Patil HS, Soman SN. Experimental Study on the Effect of Welding Speed and Tool Pin
Profiles on AA6082-O Aluminium Friction Stir Welded Butt Joints. Int J Eng Sci
Technol. 2010; 2 (5) : 268 275.
[46] Zhang Z, Zhang HW. Numerical Studies on the Effect of Transverse Speed in Friction
Stir Welding. Mater Design. 2009; 30 : 900 - 907
[47] Cui S, Chen ZW, Robson JD. A Model Relating Tool Torque and Its Associated Power
and Specific Energy to Rotation and Forward Speed During Friction Stir
Welding/Processing: Inter J Mach Tools Manuf. 2010; 50 : 1023 1030.
[48] Buffa G, Hua J, Shivpuri R, Fratini L. Design of Friction Stir Welding Tool Using the
Continuum Based FEM Model. Mater Sci Eng A. 2006; 419 : 381 388.
[49] Hattingh DG. Blignault C, van Niekerk TI, James MN. Characterization of the
Influences of FSW tool Geometry on Welding Forces and Weld Tensile Strength Using
an Instrumented Tool. J Mater Process Technol. 2008; 230 : 46 57.
[50] Arora A, De A, DebRoy T. Toward Optimum Stir Welding Tool Shoulder Diameter:
Scripta Mater. 2011; 64 : 9 12.

Principles and Thermo-Mechanical Model of Friction Stir Welding 215
[51] Frigaard , Grong , Midling OT. A Process Model for Friction Stir Welding of Age
Hardening Aluminum Alloys. Metall Mater Trans A. May 2001; 32 (5) : 1189-1200.
[52] Buffa G, Hua J, Shivpuri R, Fratini L. A Continuum Based FEM Model for Friction Stir
Welding Model Development. Mater Sci Eng A. 2006; 419 : 389 396.
[53] Nandan R, Roy GG, Lienert TJ, DebRoy T. Three-Dimensional Heat and Material Flow
During Friction Stir Welding of Mild Steel. Acta Mater. 2007; 55 : 883 895.
[54] Ulysse P. Three-Dimensional Modeling of the Friction Stir-Welding Process. Int J Mach
Tools Manuf. 2002; 42 : 1549-1557.
[55] Colegrove PA, Shercliff HR, Zettler R. Model for Predicting Heat Generation and
Temperature in Friction Stir Welding From the Material Properties. Sci Techno Weld
Joining. 2007; 12 (4) : 284-297.
[56] Chen, CM, Kovacevic, R. Thermomechanical Modeling and Force Analysis of Friction
Stir Welding by Finite Element method. In: Proceeding of the Institution of Mechanical
Engineers, Part C. J Mech Eng Sci. 2004; 218 : 509-519.
[57] Hamilton C, Dymek S, Sommer. A Thermal Model for Friction Stir Welding in
Aluminum Alloys. Int J Mach Tool Manuf. 2008; 28 : 1120-1130.
[58] Chen CM, Kovacevic R. Parametric Finite Element Analysis of Stress Evolution During
Friction Stir Welding. In: Proceeding of Institute of Mechanical Engineers Part B. J
Manuf. 2006; 22 (28) : 1359 1371.
[59] Fratini L, Buffa G, Palmeri D, Hua J, Shivpuri R., Material Flow in FSW of AA7075 T6
Butt Joints: Numerical Simulations and Experimental Verifications. Sci Technol Weld
Joining. 2006; 11 (4) : 412 421.
[60] Zhang W, Zhang Z, Chen JT. 3D Modeling of Material Flow in Friction Stir Welding
Under Different Process Parameter. J Mater Process Techn. 2007; 183 : 62-70.
[61] Pew JW, Record JH, Nelson TW, Sorensen CD. Development of A Heat Input for
Friction Stir Welding. In: ASM proceeding of International Conference: Trends in
Welding Research. 2005: 247 251.
[62] Pew JW. A Torque-Based Weld Power Model for Friction Stir Welding [Msc thesis].
Brigham Young University; December 2006.
[63] Colligan KJ, Mishra RS. A Conceptual Model for The Process Variables Related to Heat
Generation in Friction Stir Welding of Aluminum. Scripta Mater. 2008; 58 : 327 331.
[64] Benedyk JC: Aerospace Structural Metals Handbook Nonferrous Alloys; CINDAS
LLC. 2008: 68.
[65] Jauhari TK, Indra PA, Zuhailawati H. Mathematical Model for Multi Component Forces
& Torque Determination in Friction Stir Welding. Adv Mater Res. 2011; 230-232 : 1255 -
1259.
[66] Incropera, DeWitt: Fundamentals of Heat and Mass Transfer, 2th Edition; Singapore;
Wiley. 1990.
[67] Midling OT, Grong . A process model for friction welding of Al-Mg-Si alloys and Al-
SiC metal matrix composites; 1: HAZ temperature and strain rate distribution. Acta
Metall Mater. 1994; 42 : 1595-1609.
[68] Crossland B. Friction Welding. Contemporary Physics. 1971; 12 (6) : 559-574.

Welding Processes 216
[69] Longhurst WR, Strauss AM, Cook GE. Enabling Automation of Friction Stir Welding:
The Modulation of Weld Seam Input. J Dyn Sys Meas Control. 2010; 132: 1-11.
[70] Davis TA, Shin YC, Yao B., Observer-Based Adaptive Robust Control of Friction Stir
Welding Axial Force. IEEE-ASME Trans Mechatron. 2011; 16 (6) : 1032-1039
Chapter 10




2012 Soul and Hamdy, licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Numerical Simulation of Residual Stress and
Strain Behavior After Temperature Modification
Farag Soul and Nada Hamdy
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/47745
1. Introduction
Welding, among all mechanical joining processes, has been employed at an increasing rate for
its advantages in design flexibility. In addition to that, cost savings, reduced overall weight
and enhanced structural performance. The highly localized transient heat and strongly non-
linear temperature fields in both heating and cooling processes cause non-uniform thermal
expansion and contraction. Thus, result in plastic deformation in the weld and surrounding
areas. As a result, residual stress, strain and distortion are permanently produced in the
welded structures. This is particular when fabrication involves the use of thin section sheet
materials, which are not inherently stiff enough to resist the contraction forces induced by
welding. Transient thermal stresses, residual stresses, and distortion sometimes cause cracking
and mismatching of joints. High tensile residual stresses are undesirable since they can
contribute in causing fatigue failure, quench cracking and stress- corrosion cracking of welded
structures under certain conditions. Welding deformation is undesirable owing to the decrease
in buckling strength and injures the good appearance of structures.
In addition, it causes defaults during the assembly which result in repeating the process and
productivity restriction. Correction of unacceptable distortion is costly and in some cases,
impossible. In welding design, the study and analysis of welding residual stresses and
distortion become necessary in critical industries such as: aerospace engineering, nuclear
power plants, pressure vessels, boilers, marine sector.etc. Measurement of transient
thermo-mechanical history during welding process is of critical importance, but proves to be
prohibitively expensive and time consuming. It often fails to provide a complete picture of
temperature and stress/strain, deformation distribution in the weldment. On the other hand,
detailed experimental measurements of the residual elastic strain distributions in welded
parts are typically not feasible due to significant resource (man, machine and material)
consumption. Mathematical modeling for residual stress evaluation provides a resource

Welding Processes 218
effective method in comparison to the experimental methods when all interaction fields
were correctly described in the modeling process. However, development of the modeling
scheme gain demands a careful experimental data. The purpose of this chapter is to develop
Finite Element models that satisfy the analysis of the behavior of transient phenomena of
residual stress and distortion. That can be achieved by using different methods of the
mitigation technique which work as heat transfer enhancement. Approximating the
mechanisms of the transient temperature and longitudinal residual stress after temperature
modification can be made. The modeled welding materials are aluminum and titanium
alloys concerning flat and cylindrical shapes.
2. The origin of residual stress
Residual stresses developed during most manufactured processes involving metal forming,
heat treatment and machining operations deform the shape or change the properties of a
material. They arise from a number of sources and can be presented in the unprocessed raw
materials, and can be introduced during manufacturing or can arise from in-service loading.
(Withers & Bhadeshia, 2000; Rudd, 1992; Borland, 1994; Kandil et. al. , 2001 ). The residual
stresses may be high enough to cause local yielding and plastic deformation on both
microscopic and macroscopic level, that can severely affect component performance. For
this reason it is vital that some knowledge of the internal stress state can be deduced either
from measurements or modeling predictions. Both magnitude and distribution of the
residual stress can be critical to the performance that should be considered in the design of a
component. Tensile residual stresses in the surface of a component are generally undesirable
since they can contribute to the major cause of fatigue failure, quench cracking and stress-
corrosion cracking.
Compressive residual stresses in the surface layers are usually beneficial since they increase
fatigue strength, resistance to stress-corrosion cracking, and increase the bending strength of
brittle ceramics and glass. In general, residual stresses are beneficial when they operate in
the plane of the applied load and are opposite in sense (i.e, a compressive residual stress in a
component subjected to an applied tensile load). The origins of residual stresses in a
component may be classified as: mechanical, thermal and chemical. Mechanically generated
residual stresses are often a result of manufacturing processes that produce non-uniform
plastic deformation. They may develop naturally during processing or treatment, or may be
introduced deliberately to develop a particular stress profile in a component (Brien, 2000).
Examples of operations that produce undesirable surface tensile stresses or residual stress
gradients are rod or wire drawing (deep deformation), welding, machining (turning,
milling) and grinding (normal or harsh conditions). On a macroscopic level, thermally
generated residual stresses are often the consequence of non-uniform heating or cooling
operations. The residual thermal stresses coupled with the material constraints in the bulk of
a large component can lead to severe thermal gradients and the development of large
internal stresses. An example is the quenching of steel or aluminum alloys, which leads to
surface compressive stresses, balanced by tensile stresses in the bulk of the component.

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 219
Microscopic thermally generated residual stresses can also be developed in a material
during manufacture and processing as a consequence of the CTE mismatch between
different phases or constituents. The chemically generated stresses can develop due to
volume changes associated with chemical reactions, precipitation, or phase transformation.
Chemical surface treatments and coatings can lead to the generation of substantial residual
stress gradients in the surface layers of the component. Nitriding produces compressive
stress in the diffusion region because of expansion of the lattice and precipitation of nitrides
also carburizing causes a similar effect (Littmann, 1964).
3. Sheet metal fabrication
In recent years the new vision of high-tech industrial strategy looking for minimizing the
cost and increasing the strength to weight ratio of critical structure such as aerospace,
marine, nuclear etc. Thin walled element fabricated by welding process can promote such
effect. In the last two decades the research in welding science became more vital than other
manufacturing sciences in many industrial sectors. The development in welding technology
is vastly increased, and the need for sheet metal fabrication by welding is necessary for
many applications. Such those applications are rockets fuel tank and aircraft exhaust and
engine mounts). Typical example of some components used in industries are shown in
figure 1.

Figure 1. Typical thin welded component used in aerospace: (a,b,c) type of rocket fuel tank.
Other examples of sheet metal welding applications are in ships and airplanes structures.
Welding can be successfully alternated to other connection processes such as riveting. Riveting
has a low joint efficiency, thus a structure designed to be riveted, generally requires more
materials even if joint itself is not complex. The area adjacent to the rivets is also a site of high
(a)
(b) (c)

Welding Processes 220
residual stresses and stresses concentration, generating a favorable environment for stress
corrosion and initiation of fatigue cracks. Welding is considered a challenge for replacing
riveting in the future for airbus industries. The advantage of welding compared to the
mechanical fastenings, includes cost, time and ease of fabrication. Also, simpler, lighter
component design, and better joint efficiency. Laser beam welding may become a new joining
technique for aircraft fuselage shells for the A318 and A380 airbuses (Airbus, 2000). Laser
beam welding of the longitudinal stiffeners of the skin panels of a commercial aircraft fuselage
may reduce the weight of the panels to 80% against the riveted structure production Figure 2.
Welding is performed simultaneously from both sides of the stiffener. The main problem with
welding is to keep the distortion as low as possible (especially the transverse deflections) and
to reduce the residual stresses (especially the longitudinal tensile stresses).

Figure 2. The fuselage structure

Figure 3. Typical distortion on welded sheet: longit. welds (a,b,c), circular welds (d,e), and
circumference welds (f,g).

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 221
The twin problem of stress and distortion due to welding, using conventional fusion welding
process, have presented fabrication problems for many years especially, in aerospace industry.
Consequently, manufacturing have applied additional time consuming and costly operations.
This is to remove distortions or to relieve stresses after welding that avoid variable quality
problem. This is particularly so when fabrication involves the use of thin sheet section,
typically in the range of 0.5-4 mm in thickness. Recently advanced aluminum and titanium
alloys became very attractive materials in sheet metal fabrication in high tech- industries due
to their strength to weight ratio. However, as the thickness decreases, the sheet materials are
not stiff enough to resist the contraction forces induced by welding. In the aerospace
fabrication, the longitudinal residual stress in addition to distortion such as longitudinal
bending and buckling become more substantial. The manufacturing of welded cylinders,
cones or other shaped shell elements in aerospace industries are always accompanied by
distortions. Figure 3, shows typical example of those distortions produced by longitudinal or
circumferential welding when fusion welding process is applied (Guan, 1999).
4. Welding and its focus research
Historically, electric arc welding appeared in the late 19
th
century, shortly after electric power
became available. Other fusion welding processes were recently developed such as electron
beam welding (EB) and laser beam welding (LB), which introduced new generations in the
welding equipments and processes. Failure of welded bridge in Europe in the 1930 and the
American liberty ships in world war II make the concern of welding mechanics is important.
The welding researches carried out since that time, then vastly developed. Analyses of these
subjects require complex computation; therefore, most early studies were primarily empirical
or limited to the analysis of simple cases. A number of studies have been performed on the
calculation of residual stress and deformation, but few are useful in the design process. Most
of these calculation methods are limited for special purposes, e.g. (Hansen, 1968), or too
complicated to be used in design and production, e.g. (Okerblom, 1955). (Puchaicela, 1997),
exhibit some empirical formulas for general distortion modes in welded steel structures. Most
of the formula are based on measurements of deformation and strains and cannot take in to
account what really happens in the (HAZ) when the welding is cooling down.
Attempts were made to investigate changes in the bending deformation due varying
pertinent parameters, see (Masubuchi, 1980). The investigation cover many practical aspects
of process, but based mainly on experimental data. It is difficult to analyze the process,
which is highly non-linear and involves plastic deformations and high temperatures varying
in both time and space. Several reviews are, however, available such as the extensive review
done by (Masubuchi, 1980; Radaj, 1992; Goldak et al., 1992) and the recent one by (Lindgren,
2001a, 2001b, 2001c). With the advancement of modern computers and computational
techniques (for example, the finite-element and finite-difference method), a renewed effort
has been made in recent years to study residual stresses and related phenomena. Therefore,
it is now possible using computer programs to simulate the transient thermal stresses and
metal movement during welding, as well as, the residual stresses and distortion that remain
after welding is completed as found by (Tall, 1991 ; Hibbit & Marcel, 1972; Muraki et. al.,
1975; Rybicki et. al., 1978 ).

Welding Processes 222
4.1. The finite element method in historical perspective
The history of the finite element method is about hundred years, but it took another fifty
years before the method became useful. In 1906 a paper was presented where researchers
suggested a method for replacing the continuum description for stress analysis by a regular
pattern of elastic bars. Later (Courant, 1943) proposed the finite element method, as we
know it today, the residue of section cover the developed FEM during the years and the
motivation computer tools based on FEA, and the use for those packages for simulation of
welding phenomena by the researcher. Many authors have utilized the commercial finite
element codes ABAQUS & ANSYS enhanced with user subroutines, to model weld
simulations with great success (Dong et. al., 1998; Feng et. al.,1996; Karlsson et. al., 1989;
Grong & Myhr, 1993; Voss et. al., 1999; Tenga & Linb, 1998; Li et. al.,2004). The finite element
code ADINAT was used by (Karlsson & Josefson, 1990), while other authors (Junek et. al.,
1999; Vincent et. al., 1999; Dubois et. al., 1984) have utilized SYSWELD to perform weld
simulations. Welding is complex industrial process which often requires several trials before
it can be done right. The welding is carried out by skilled workers, but in the past few years
automated machines and robots are sufficiently used in the small and large industrial scales.
To obtain the expected productivity through mechanization, high precision of the assembled
parts must be kept. Therefore, the predictability is important in such aerospace, shipyards,
nuclear and automobile industries. In order to produce a high-quality product, the accuracy
control should be kept through the whole assembly line. The concept of accuracy control
should be incorporated in the structural design, so that the designer can produce a better
design accounting for the geometric inaccuracy. Numerical modeling and simulation of
welding are a difficult and challenging problem due to the complex mechanisms involved.
The wide range of problems concerned can be generalized into the fields shown in figure 4.
The fields are strongly interrelated and couple in almost every possible manner.
Establishment of a model accounting for all the physical effects and their couplings would
be an incomprehensibly large and complex task. Hence, welding research is characterized
by choice of a focal area for thorough analysis and use of suitable assumptions. Thus, the
'art' of welding research is to choose simplifications without invalidating the research focus.

Figure 4. Coupled fields in welding analysis.

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 223
4.2. Welding induced residual stresses
Stresses arising during the welding process are referred to as internal or locked-in stresses
(Radaj, 1992; Gatovskii & Karkhin, 1980). Residual stresses can be defined as those stresses that
remain in a material or body after manufacture and processing in the absence of external forces
or thermal gradients. Residual stress measurement techniques invariably measure strains rather
than stresses, and the residual stresses are then deduced using the appropriate material
parameters such as Youngs modulus and Poissons ratio. Often only a single stress value is
quoted and the stresses are implicitly assumed to be constant within the measurement volume,
both in the surface plane and through the depth. Residual stresses can be defined as either
macro or micro stresses and both may be present in a component at anyone time. Macro
residual stresses, which are often referred to as Type I residual stresses, vary within the body of
the component over a range much larger than the grain size. Micro residual stresses, which
result from differences within the microstructure of a material, can be classified as Type II or III.
Type II residual stresses are micro residual stresses that operate at the grain-size level; Type III
are generated at the atomic level. Micro residual stresses often result from the presence of
different phases or constituents in a material. They can change sign or magnitude over distances
comparable to the grain size of the material under analysis. To summarize, Residual stresses in
the material body can be classified three type as found in (Withers & Bhadeshia, 2000; Borland,
1994; Noyan). The different types of residual stress are shown schematically in figure 5.

Figure 5. Categorization of residual stresses according to length scales.
Welding stresses can be classified by three characteristics: By lifetime, welding stresses can
be temporary or residual, the temporary stresses do exist only in a specific moment of the
non-stationary process of heating and cooling. The residual stresses can be found after the
whole process of welding is completed and structure is cooled down to the room
temperature. By directional the welding stresses subdivide into longitudinal (parallel to the
welding direction) and transversal (perpendicular to the weld seam) and through thickness
stress. By the origins, the welding stresses are subdivided into. Thermal stress, Stresses
caused by the plastic deformation of the metal; Stresses caused by phase transformations.

Welding Processes 224
4.2.1. Welding induced longitudinal residual stress
Representation of the temperature and the resulting longitudinal stress distributions that
occur during welding are schematically gives in figure 6. In this example a simple bead-on-
plate case is analyzed (see figure 6a). The welding arc which is moving along the x-axis with
a speed v, is indicated by the arrow. Far ahead from the heat source the temperature is
constant and the stress is equal to zero in all the points. Moving in the negative direction of
the x-axis, we reach the point where the temperature starts to rise figure 6c. The points close
to the weld line start to experience compression in the longitudinal direction. This deep fall
changes to a fast rise of the longitudinal stress. The rate of stress change is proportional to
the temperature gradient ahead of the source. It caused by the yielding point yield
changing with temperature. As known, at elevated temperatures the material begins to
soften. After some temperature (the softening temperature) the material reaches the stage
when Y is almost zero, and so, the points situated close to the centerline reach the softening
temperature, and climb up to a zero value of the longitudinal stress. Stresses in the regions a
short distance from the arc are compressive, because the surrounding metal restrains the
expansion of these areas where the temperature is lower. However, stresses in the areas
further away from the weld arc are tensile and balanced by compressive stresses in the areas
near the weld. Going further, at some distance behind the welding arc, the temperature
drops sufficiently for the material to be stiff enough to resist the deformation caused by the
temperature change. Due to cooling the areas close to the weld contract and cause tensile
stresses. After a certain time, the temperature change due to welding diminishes. High
tensile longitudinal stresses (usually up to the yielding stress) are produced near the weld.
In the regions further away from the weld, compressive stresses do exist. Figure 6d describe
the final distribution of longitudinal residual stress, from literature (Masubuchi, 1980), X
can be approximated by:

2
[1/2( / ) ] 2
( ) [1 ( / ) ]
y b
x m
y y b e

(1)
Where M is the maximum stress at the welding line, y is the distance from the weld line, b
width of tension stress.

Figure 6. Schematic representation, (a, b, c) temperature vs stress during welding[Pilipenko]; d) final
longitudinal residual stress.

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 225
4.3. Welding induced deformation
As in case of the stresses occurring during and after welding, welding deformation can be
transient or residual. Figure 7 gives an overview of various types of welding deformations
to be expected when welding plates.

Figure 7. Different types of welding distortions. The arrows indicate the shrinkage direction of the weld
metal which causes the corresponding distortion [Masubuchi 1980].
All these kinds of distortions are related to the shrinkage of the weld metal during cooling.
They can be subdivided into:
1. Transverse shrinkage shrinkage perpendicular to the weld seam
2. Longitudinal shrinkage shrinkage in direction of the weld seam
3. Angular distortion transverse uplift caused by a non-uniform temperature
distribution in the through-thickness direction. For instance in case of butt-joints with a
V-groove.
4. Rotational distortion in-plane angular distortion due to the localized thermal
expansion and contractions. Very relevant for overlap joints, for instance.
5. Bending distortion longitudinal uplift. The same causes as angular distortion.
6. Buckling distortion caused by compressive stresses inducing instabilities in the plates.
Driven by the need to save fuel and reduce transport and operating costs, there is a
growing demand for lightweight structures, for example in the automotive and aircraft
industries as well as in shipbuilding. At the very basis of this trend we find the
availability of recently developed metallic alloys that actually allow the transition to more
lightweight designs. Although many of the welding techniques that are currently
available offer suitable material and mechanical properties, the degree of distortion
remains unacceptable and residual stresses often approach component design limits. The
increasing reduction in thickness will lead to a growing demand for effective solutions for
residual stress and strain control during welding. An example of welding deformations in
thin sheet structures can be found in the shipbuilding industry, where welding causes a
typical wave-like appearance on the hull of a ship (see Figure 8). Such problem results

Welding Processes 226
through various stages of production have emerged as a major obstacle to the cost-
effective fabrication of lightweight structures. Same situation may occurs in aerospace and
aircraft assembly where the high strength to weight ratio are necessary and thin elements
are used, in addition to the requirement of smooth surface to maximize hydrodynamic
performance and minimize radar signature [Huang 2004]. A conservative estimation for
the labor costs accumulating for post-welding distortion correction is approximately 30 %
[Andersen 2000; van der Aa,2007].

Figure 8. Ship hull defect due to distortion (van der Aa, 2007) .
4.3.1. Bending distortion in welded sheet metal
Bending distortion in sheet metal can be schematically shown in figure 9. When structure is
welded, heat is supplied to melt the joint and non-uniform temperature distribution is
caused owing to local melting, As a result, non-uniform thermal strains and stresses are
caused and plastic strains remain after thermal cycle. Residual plastic strain around welded
joints is the cause of permanent deformation. Figure 9a presents the pure cambering which
may occur when the ratio of sheet length to width is high enough, but when this ratio
decreases, the sheet exposes both cambering and angular distortion as shown in figure 9b. In
small thickness, angular deformation is not significant because of the high homogeneity of
the temperature field through the plate thickness. In some cases gradient forces countered
from longitudinal shrinkage Fx, as shown in figure 9c, are more dominant and cause the
cambering owing to low stiffness of thin sheet at this moment. This may occur when the
longitudinal residual stress above the neutral axis of the sheet exceeds that below the
neutral axis.
The possibility for minimizing or eliminating this problem is only to balance the
longitudinal stress around the neutral axis otherwise, minimizing these stresses below the
significant magnitude which not exceeds the component stiffness. For most welding
processes, the incident surface will absorb the most energy, with the energy absorption
decreasing with depth. The variation in the through-thickness heating causes variation in
the longitudinal stresses through the plate thickness. This generates a bending moment,
which causes the bending distortion mode.

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 227




Figure 9. Schematic of: (a) cambering; (b) combined cambering and angular; (c) Shrinkage foresees in
inherent strain region.
4.4. Novel control techniques for residual stress and distortion
Controlling of distortion has been investigated in series of papers by the Edison welding
institute (EWI) (Conrady & Dull, 1995; Michaleris et al., 1999; Michaleris & Sun, 1997;
Conrardy & Dull, 1997). Those papers concerned for the thermal tensioning technique in
both static and dynamic heating during welding processes. The technique was found to be
an active method in welded ship structures. Different heating sources can be used
enhancing tensioning effect such as dynamic flame heating and moving laser spot heating or
static heating. Another mitigation technique for controlling welding-induced stresses and
distortion has been developed by Beijing Aeronautical Manufacturing Technology Research
institute (Q. Guan et. al., 1994; Guan et al., 1993). The technique called dynamic controlled
low stress no distortion (DC-LSND); it has been applied successfully to aerospace
manufacturing for shell structures such as jet engine cases of heat resistance alloys and
rocket fuel tanks of aluminum alloys (Guan et al., 1993; Guan et al., 1996). Many literature
present that the residual stress can be minimized by using (DC-LSND) welding technique (
Li et al , 2004a; Li et al., 2004b). many cooling media can be used in this technique such as
(atomized water, compressed air, solid CO2, liquid nitrogen, liquid argon). Beside the
reduction of plastic strain, it was found that heat transfer enhancement by trailing heat sink
technique work as source for balancing residual stresses above and under the neutral axis
(Soul & Yanhua, 2005, 2006; Soul et al, 2010). The set-up of dynamic heating spots and
trailing heat sink are represented schematically in figure 10. The studies on the temperature
field characteristics and the thermal history are the foundation and prerequisite to study the
stress and distortion control mechanism in welding mitigating techniques. Sometimes, it is
inconvenient or even impossible to obtain the real thermal cycle at weld pool by experiment
due to its limitations.
(c)
(a)
(b)

Welding Processes 228

Figure 10. Schematic drawing represents: (a) moving heating spots; (b) trailing cooling spot
To overcome this disadvantage, the advanced numerical analysis technologies, such as finite
element method and finite difference method, have been frequently used to obtain the whole
temperature field of the welded specimen. It is necessary to develop a computer-based tool to
optimize welding mitigation processes and hence minimize the expense and time incurred by
extensive welding trials. In this chapter three dimensional finite element methods are employed
to find qualitative analysis for the temperature field, residual stress and transient plastic strain
developed during welding process. To mitigate the problem of residual stress and distortion,
two different techniques were tested and compared, those are trailing heat sink and dynamic
heating spots techniques. Gas tungsten arc welding process (GTAW) is used for simulation.
5. Investigation of residual stress behaviour after enhanced heat transfer
5.1. Trailing cooling spot
The proposed welding technique incorporates a trailing heat sink (an intense cooling source)
with respect to the welding torch, and it is also named Low Stress No Distortion (LSND)
welding. The development of this mitigation technique is based on both detailed welding
process simulation using advanced finite element method and systematic laboratory trials.
For understanding well LSND welding, finite element method is used to investigate the
mechanism of the technique. In this chapter, 3D-FEA results from different papers done by
the author have been selected. These results study the mechanism of the trailing heat sink
mitigation technique and how the longitudinal residual stress was minimized based on the
distance between the torch and the cooling spot. The qualitative and quantitative analysis of
residual stress depend on the temperature gradient in the component during heating and
cooling. So any modification of temperature topography may decrease or increase the
residual stress that depends on interaction of strains. Three-dimensional models for
welding, the thermal cycle and residual stress in welding are now in common use as a
research tool for both academic and commercial purposes. The models use a transient 3-
dimensional thermal model which is decoupled to an elastic-plastic model for calculating
the stress and strain. Models were investigated with different material and dimension as
shown in table 1.
(a)
(b)

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 229
Model Size geometry
Model 1 Al-Mn(3003) 260 130 2 mm mm mm Flat sheet
Model 2 Al-Mg (5083) 240 80 3 mm mm mm Flat sheet
Model 3 Ti-6Al-4V 270 120 2.5 mm mm mm Flat sheet
Model4 AL-Cu (2024 100mmOD, 96mmID, 240mm length Cylindrical sheet
Table 1. Simulated models: material, shapes and their dimensions.
5.1.1. Establishment of heat source model
In this study, two heat source models were investigated, then correlated on fitting a
practical welded sample for distinguishing the better one, so it can be used to detect other
analysis. The first one is disc model proposed by (pavelic et al, 1969 ), the mathematical
expression of the model present in Equation 2.


2
2
0
3
2
0
3
r
r
Q
q r e
r

(2)
Where q(r) is the surface flux at radius r (W/m
2
), r0 is the region in which 95 % of the heat
flux is deposited, r is radial distance from center of the heat source and Q is the heat
input.
The second model is double ellipsoidal power density distribution" adopted from (Goldak,
1984), the mathematical expression of the model present in Equations 3 and 4.

2 2
2 2 2 2
( 3 / )
( 3 / ) ( 3 / )
6 3
( , , )
f
z c f
y b x a
f
f
Qf
q x y z e e e
abc

(3)

2 2 2 2 2 2
( 3 / ) ( 3 / ) ( 3 / )
6 3
( , , )
r
z c y b x a r
r
r
Qf
q x y z e e e
abc

(4)
Where a, b, c are the semi-axis for the gaussian distribution in (x,y,z) direction respectively
and ff, fr are fractions of heat deposit in front and rear of the heat source. The intensity plot
for both surface heat and double ellipsoid heat source models is shown in figure 11.
Variation of the semi-axis and the heat deposit fractions allows the double ellipsoid fitted to
give suitable heat source especially at increased welding speed.
At the same heat input, the double ellipsoid is more reasonable for fitting the fusion
boundary rather than the gaussian distribution or surface heat source model as depicted in
figure 12.

Welding Processes 230

Figure 11. Intensity of heat source model: (a) gaussian distribution; b) double ellipsoid

Figure 12. Experimental & simulated fusion boundary fitting: (a) double ellipsoid; (b) gaussian
distribution;
5.1.2. Effects of trailing cooling spot on heat transfer form
The temperature distribution resulted from the thermal analysis for the heat sink process stated
above is predicted at a welding time of 25s and presented in Figure 13. The temperature
decreases drastically in the zone between the arc and the heat sink, and the corresponding
temperature gradient increases. Further more, in the front of the heat sink the temperature
isotherms reveals the existence of high temperature gradient and therefore, some of high
temperature contours are drawn back to the front of heat sink and distorted temperature
distribution were formed. However, the point in and near the weld centerline may passed by
more than one thermal cycle and this may expressed as [heating - normal cooling - forced
cooling surrounding heating normal cooling]. The evidence of those cycles became more
clear when different locations in the samples in both conventional and after applying heat sink
processes were selected. Figure 14 reveals different location with difference cycle profiles.
The locations at or near the weld line present those cycles and exhibit significance difference
in thermal profiles. The point at weld line reveals the maximum intensity of the cooling spot
which indicate a steep profile and the existence of valley . The thickness of the modeled
sample play an important role for the effectiveness of the technique. Figure 14c shows the
thermal history of simulated welded aluminum alloy with 2mm in thickness
(a)
(b)
(a)
(b)

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 231








Figure 13. Temperature contours in trailing cooling spot: a) Al-alloy; b) Ti- alloy (soul, 2oo5,2006)








Figure 14. Thermal history: (a) conventional welding; (b) trailing cooling spot[t=3mm], (c) [t=2mm]
Comparing these results with that modeled with thickness 3mm as presented in figure 14b,
at the point (y=0), the penetration of cooling zone is more effective in the small thickness,
which reveals lower temperature magnitude at the center of the spot figure 14c. Moreover,
(a) (b)
(a) (b) (c)

Welding Processes 232
the overall temperature tips in thermal history is decreased compared with conventional
welding. This refers to some energy absorbed from the total energy by the amount action of
introduced cooling.
5.1.3. Longitudinal residual stress behaviour
The result obtained from the 3-D modeling of welding a bead on plate with thickness 2.5mm
for titanium alloy and 3mm for aluminum alloy using GTAW process were depicted in
figure 15. Describing the process behavior, in the front of the torch, compression transient
residual stress were developed which result from generated thermal strain. When the
material loose the mechanical properties at high temperature such yield strength and
young

s modulus, no extra stresses were produced. Behind the welding pool when the
shrinkage started , plastic strain was accumulated and transient residual stress changes from
compression to tension in short period of time. Therefore, Its magnitude increased vastly as
the temperature decreased. However, it is an important knowledge if this change with
duration can be measured practically to make correlation, but it is so difficult. In fact, the
final magnitude of residual stress depends on cooling rate, regarding the metallurgical
science welding microstructure morphology and size depend on the cooling rate as well. So
no welding residual stress developed if there is no change or homogenous microstructure
was obtained.

Figure 15. Transient residual stress: (a) Al-Mg alloy; (b) Ti-6Al-4V alloy
In more details, after including trailing heat sink technique, which can enhance heat transfer
process, the temperature was modified. The analyzed results in figure 16 obtained when the
trailing heat sink is located at 30 mm behind the torch, exhibit that the stress profile has
different behavior. For instance, in small area depicted in figure 16a (zone of compressed
contours), the contours behind the torch were compressed in the front of the cooling spot,
and its shape becomes in complicated form through this stage.
(a) (b)

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 233

Figure 16. Transient longitudinal residual: (a) stress vs thermal cycle ; (b) longitudinal residual stress
contours Al-Mn alloy
The transient longitudinal residual stress stays constant for a period of time, this may also
be referred to the high temperature contours intercepted between the two sources which
dont affect the plastic strain. Through enough length along weld centerline behind cooling
zone, the longitudinal residual stress seams to have a constant value which can be
distinguished from a stable color along this length. This behavior coupled with the constant
values of contours around mentioned length.
Moving a step back, far from local cooling zone, the temperature of the metal at the spot-
cooling region is smaller than the surrounding, so the fast contraction of welded metal in the
cooling zone and high transient longitudinal residual stress may developed as depicted in
figure 16b (red color were the arrow located). Because of cooling that crossed in a short time
and the hot contours appeared again due to the hot surrounding material, the residual plastic
strain may released and recovery process may generate. However, both the effectiveness of the
cooling at the upper surface and the abnormal temperature developed behind the heat source
may affect the balance level for the front-to-rear stress pattern. Furthermore, in short distance
behind the torch, high temperature contours around the cooling zone and low temperature
inside, which revealed a big difference from that occurred in conventional welding process.
Therefore, it may brought less stress during solidification temperature range, and reasonable
longitudinal strain can be obtained. Behind the cooling zone the transient longitudinal residual
stress profile seems to be decreases to somewhat value due to an increase in temperature as
shown in figure 17. In this stage, the process became as heat treatment for the residual stress,
and the metal may expand again due to heating which believed to be appeared in the
expansion process in the rear of the cooling zone.
(a) (b)

Welding Processes 234
The drop in residual stress magnitude follows the change in temperature according to the
thermal cycle may refer to the opposite change of the elastic strain when the temperature
increases again by the surrounding hot metal. Therefore, the reasonable expression for this
mechanism related to recovery process due to the heating process behind the cooling spot
which may expand again, and then some of the plastic strain can be released. In practice,
this process is quite logic during the heat treatment of materials suffering from residual
stress induced by welding or other strengthening process such as cold working etc.
Furthermore, the maximum longitudinal residual stress in titanium model is reduced to
about (~326 MPa) from that in conventional welding, where in the aluminum model the
maximum residual stress is minimized about (~55MPa) from that in conventional welding.
The tested technique shows the significance influence on the final longitudinal residual
stress through thickness, the clear evidence for such influence can be seen in figure 18. The
residual stress at the upper surface became less in magnitude than that in the middle of
plate thickness. It seems not only the over all stress minimization but the possibilities for
balance the stresses around the center of the Thickness.












Figure 17. Transient longitudinal residual stress in moving cooling spot: (a) in Al-Mg alloy; (b) in Ti-
6Al-4V alloy
(a) (b)

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 235





Figure 18. Final longitudinal Residual Stress through Thickness: (a) In Conventional Welding; (b) In
Trailing Heat Sink.
5.1.4. Longitudinal plastic strain behavior
In welding process, the residual stress magnitude based on the inherent strain or plastic
strain, so the analysis of strain behavior is important to characterize the minimization of
residual stress. In the work done by (soul & yan hua,2006) the transient longitudinal plastic
strain profiles were analyzed at two nodes and presented in figure 19. In conventional
welding the plastic strain at the centerline (Y=0mm) seams to be higher than the point a little
moved a way from the center line (Y=2.5mm). However; after including dynamic cooling
spot figure 19b, significant change in plastic strain behavior was obtained. The plastic strain
at the centerline became lower than that at (2.5mm), it is clear that the overall plastic strain
magnitude were minimized due to the used technique.
The minimization behavior refers to the complex tensioning effect based on the
temperature topography. Moreover, due to the surrounding hot material, the temperature
increased again behind the trailing cooling spot, and some of strains is released or partial
annealing process occurs. In practical welding, it is difficult to obtain constant
temperature through thickness with respect to the time whatever it is thin. This non-
homogeneity refers to the three heat transfer conditions which may occurs at the upper
surface rather than at the lower surface, hence the temperature is still higher above. To
characterize the change in plastic strain through the thickness, simulated result of 2.5 mm
modeled welded titanium sheet is analyzed. However, when no backing used in welding,
the cooling rate at upper surface is still more than at lower surface. Figure 20 reveals that
the maximum plastic strain occurs at the upper surface but after trailing heat sink
introduced figure 20b, all the plastic strain magnitude through the thickness became close
to each other due to the process effects.
(a) (b)

Welding Processes 236




Figure 19. Transient plastic strain: (a) in conventional welding; (b) in trailing heat sink.



Figure 20. Plastic strain through depth: (a) in conventional welding; (b) in trailing heat sink.
Moreover, the plastic strain magnitude at the upper surface is less or equal the strain at the
lower surface which cause a balance in strains around the neutral axis. Therefore, balancing
of forces occurred, and bending distortion minimized or can be eliminate depends on the
degree of balancing. It is conclude that the reduction in overall longitudinal plastic strain
magnitude is the source of longitudinal residual stress minimization. This results can be
coupled with that obtained in Figs. 15b and 17b.
(a) (b)
(a)
(b)

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 237
5.1.5. Effectiveness of the process on longitudinal welded cylinder model
In the cylindrical model the trailing heat sink considered to be moving behind the heat
source at the bottom side. This consideration according to the results obtained in
conventional welding which indicate that the maximum axial residual tensile stress
occurred at inner the surface. The temperature distribution resulted from the thermal
analysis of those conventional welding and that including trailing cooling spot welding
process stated in the above sections are studied at welding time of 20s. Contour plots of
the temperature distribution under the two conditions are shown in figure 21.
Comparing the isotherms in both process results, the contours behind the heat source
were shifted in the front of trailing cooling spot due to abnormal heat transfer process.
The homogeneous heat transfer in conventional welding looses its stability after cooling
spot was introduced. Moreover, the temperature magnitude in compressed contours was
decreased due to the absorbed heat at high temperature behind the torch by the action of
cooling. The effect of change in heat transfer mechanism on the distribution of axial
residual stress can be seen in figure 22. the residual stress profile is completely different
after introduced dynamic cooling spot. Because of the cylindrical shape is different from
the flat from point of view the stiffness resistance from point of view, the reduction in
final residual stress in cylindrical model is not sufficient when it is compared with the
flat model which is still thicker.








Figure 21. Temperature distribution at 20 s: (a) conventional welding; (b) trailing heat sink.
(a) (b)

Welding Processes 238

Figure 22. Transient axial residual stress (a) conventional welding; (b) trailing heat sink.
5.1.6. Control of distortion by trailing cooling spot
As discussed in the previous sections, how the mechanism of heat transfer was changed
after introducing heat sink. Therefore, it results in thermal tension that differs from what
occurs in conventional one. The degree of tension depends on the position of the cooling
spot with respect to the heat source, or the proper parameters as denoted in ( Li et al, 2004,
2005; Soul et. al., 2006). All the above results show the minimization in longitudinal residual
stress that refers to the decrease in responsible plastic strain. All of the previous parameters
in addition of balancing stress and strain on the upper and under the neutral axis bring the
elimination of the structure distortion as depicted in figure23 and figure 24.

Figure 23. Final distortion: (a) conventional process; (b) with heat sink process;
(c) welded experiment sample.
(a)
(b)

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 239

Figure 24. Final deformed model: (a) GTAW process; (b) DC-LSND process.
5.2. Optimization of thermal tensioning in welded shell element
There are many different methods of welding deformation and stress reduction. At the same
time, the main principles are the same. Classification of the mitigation techniques helps to
understand its performance capabilities and limitations. According to the basic mechanisms
on which the techniques are based.
5.2.1. Effect of the laser heating spot on the temperature distribution
The Introducing of heating spot besides the welding reveals an increase in temperature
at fusion zone and other near regions comparing with the conventional welding itself.
So, to obtain the similar contours of temperature at welding line, the heat input from the
torch were decreased about 19%. Figure 25 shows the predicted temperature distribution
in conventional welding and moving heating spots respectively. The selected separated
positions for laser spots were 34 mm from the centerline. It is clear, that the heating spot
not only affects the temperature distribution in the front of heat source, but also enlarges
the contours in transverse direction far way from centerline. This is due to the increasing
of the total energy input with respect to small specimen width. However the contours
change completely at the position of spots due to the concentration of laser heating.



Figure 25. Temperature distribution contours: (a) conventional welding; (b) moving heat spot.
(a)
(b)
(a) (b)

Welding Processes 240
5.2.2. Effects of the laser heating spot on the stress and strain behavior
Longitudinal residual stresses and plastic strains in transverse direction with respect to
the time are presented in figure 26. The analyzed points are taken at different distance
from the centerline in the mid length of the model. The distance of heat spot from weld
centerline is 25 mm. The tensile stress profile developed during cooling cycle show
different behavior compared to the result obtained from the conventional welding figure
26a. However, the transient longitudinal residual stress stay constant during some period
of time (~40s), Thus, at that time the heat source already switched off. This behavior is
related to the effect of spot heating. The heated material by laser spot were expanded to
the direction of welding centerline, in the same time the metal behind the torch vastly
contracted, now there are two possibilities, first, if the material is heated for enough
distance in transfer direction the contraction of welded material dont meet the resistance
from the heated material for mentioned period of time above, in the opposite of the
conventional process were the cold material restrained the contraction. Second, when the
material contacted at the welding line the heated material by spots expanded in the
direction of welding line, hence the material may contracted without more tension due to
the expanded material neighboring during the above mentioned period. Moreover the
residual stress using this method still more than in cooling spot technique. Concerning the
plastic strain behavior in this technique, the transient plastic strain result from laser
heating spots reveals complicated profile too; the longitudinal plastic compressive strain
is caused by the expansion of the heated material being constraint by the cooler material
nearby.

Figure 26. Simulated results in moving heating spots: (a) Transient residual stress; (b) Transient plastic
strain
The rising of temperature far ways from zone induced by laser spots affects the material
properties and decrease it is stiffness, hence the resistance of material to the weld metal
contraction were reduced. Moreover the transient longitudinal plastic take long period of
(a)
(a)

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 241
time to accomplished final magnitude due to increase of overall body temperature, and
its value less than from that obtained by trailing heat sink technique figure 17a, but the
residual stress and distortion is still more. This phenomenon may refer to other strains
results in transverse and through thickness direction, which need more details for their
behavior under this process due to the complicated strain interaction in the three
dimensional
5.2.3. Processes parameters controlling distortion degree
It is clear from the previous studies that any modifications in the longitudinal residual
stress results from temperature modification or heat transfer enhancement. As what
happen in the simulated techniques, and the main control on the twin problem is how the
temperature topography looks like. However, the change of processes parameters will
give changes in temperature topography. Therefore, different temperature topography
generate different tensioning process or different strain interaction. According to the
previous explanation, for the process optimization investigation of the effect of the
distance between the heating spot and weld centerline on the cambering distortion
magnitude was carried out. Figure 27 presents the degree of distortion obtained from
computational results at different distance of the laser spot. The results show that the
distortion increases as the distance increases. However, after introducing the heat sink
process as depicted in figure 27b, the distortion may disappear and the calculated
displacement is only 1.3 mm. The result is quite reasonable according to the minimized
and balanced residual longitudinal stress and plastic strain or the balanced forces and
moment which become insignificant to produce the curvature along welding line as
mentioned in the previous sections. Moreover, the displacement in laser heating spot was
reduced but still higher than that obtained in the cooling technique. At this point
comparing both mitigation techniques, it can be concluded that at the same reduction of
distortion, still the moving cooling spot more effective technique than thermal tensioning
for residual stress minimization in thin element.

Figure 27. Parameters effects the processes performance: (a) heating spots; (b) cooling spot.
0
1
2
3
4
10 20 30 40
Laser arc separation-trans.(mm)
D
i
s
p
l
a
c
e
m
e
n
t
(
m
m
)
0
1
2
3
10 20 30 40
Distance between sources(mm)
D
i
s
p
l
a
c
e
m
e
n
t
(
m
m
)
(a) (b)

Welding Processes 242
6. Conclusions
One of the major problems induced during welding of thin element structure are the twin
Problem. Residual stresses reduce the performance of welded component when they have the
same sense with the working stress in service. Other disadvantage of residual stress is
promoting the stress corrosion cracking. Distortion from other side, impairs the structure
appearance and causes misalignment during assemblies. During the years, many mitigating
techniques are developed in literature to reduce the residual stress and elimination of
distortion. In the end of last century, techniques introduced thermal tensioning or temperature
modification were creative, that including (LSND) and heating techniques. When the dynamic
action of these techniques applied, they became more suitable to approximate the solution of
the problem, because it is dealing with temperature modification, and it is known that the
temperature is the source of the problem. In this chapter, the two mentioned active mitigating
techniques were investigated in dynamic action and correlated with each other. The main
fields of interest for the investigation are temperature fields, longitudinal residual stress and
strain in addition to the bending distortion. Atomized cooling water with heat transfer
coefficient were proposed , also laser heating spot beside the torch with proper heat input
were considered. The output results of this investigation summarized as following:
The change of heat transfer enhancement produce temperature modification or
different temperature topography.
The modification of temperature distribution generate different tensioning inside the
body, therefore new abnormal strain interaction in different direction.
Due to complexity stiffness based-response for different shape, the transient stress
behavior developed in the bead on plate model show different profile correlate with
that in the cylinder model.
One of the major advantages of FE optimization approach over welding trials was that
the models enabled the transient stresses and strains during welding to be considered.
This greatly simplified the understanding and optimization process.
During the DC-LSND process, the thermal history for metal in the weld centerline and
at closed region passed more than one thermal cycles, as (heating-cooling-heating-
cooling) and the duration time for staying the metal at high temperature is shorter than
in conventional process. This is found to reveal different stress and strain history
profiles at this region compared with the profiles obtained in the conventional process.
The significant reduction in the residual stress obtained at the top of surface and the
maximum stress at the middle of the thickness. This helps to balance the shrinkage
forces above and under the weld metal center, and distortion can be prevented
The suggested techniques gave a reduction of the peak residual stress. Optimization of
both DC-LSND and thermal tensioning can give similar reduction or elimination of
bending distortion of such used models thickness. At this optimization still the DC-
LSND method has a better effectiveness on longitudinal residual stress minimization
Future work analysis need for strains in other (Y and Z) directions, to see it is behavior after
applying the above techniques and correlate with the longitudinal direction.

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 243
Author details
Farag Soul
University of Elmergeb, Faculty of Engineering, Libya
Hamdy Nada
University of Menoufia, Faculty of Engineering, Egypt
7. References
Withers P. J. & Bhadeshia H. (2000), Residual Stress II: Nature and Origins, Mat. Sci. Tech.
Rudd C. O. (199), Residual stresses and their measurement, Proc. 1st Int. Conf. on Quenching
& Control of Distortion, Chicago, Illinois, USA, 193-198.
Borland D. W.(1994), Residual stress measurement Methods, limitations and significance,
Proc. 2nd Australian Int. Conf. Surface Engineering, Adelaide, 114-121.
Kandil F. A. et. al. (2001); A Review Of Residual Stress Measurement Methods -A Guide To
Technique Selection, NPL Report MATC(A)O4.
Brien E. O. (2000), Residual Stress Friend and Foe: A Civil Aircraft Perspective, ICRS-6,
Oxford, UK, 10-12 July, vol. 1, 13-21.
Evans E. B.(1986), Residual stresses in processing. Encyclopaedia of Materials Science &
Engineering, Vol. 6, 4183-4188
Littmann W. E.(1964), Measurement and significance of residual macrostress in steel, SAE
793A, Proc. of the Automatic Eng. Cong., Detroit, MI, 13-17.
Airbus 2000, http://www.airbus.com /gmf99_1.html
Guan Q. (1999), A survey of development in welding stress and distortion controlling in
aerospace manufacturing engineering in china Welding in the world, Pergamon press
Ltd, 43(1).
Hansen B.(1968) , Welding stresses and deformation. Svejse Copenhagen report no. 68.07
edition, 1968. in Danish.
Okerblom N. O. (1955), The Calculations Of Deformations Of Welded Metal Structures. Mashgiz,
Moscow-Leningrad, 1955.
Puchaicela J. (1997), Control of distortion of welded steel structures, Welding J., pp.49-52.
Masubuchi K. (1980), Analysis of welded structures, Pergamon Press, London.
Radaj R. (1992), Heat Effects of Welding, Spring-Verlag.
Goldak J. et. al. (1992), Coupling heat transfer, microstructure evolution and thermal stress
analysis in weld mechanics, Proc. of IUTAM Mechanical effects of welding, Springer
Verlag, Berlin.
Lindgren L. (2001), Finite element modeling and simulation of welding part 1: Increased
Complexity, Journal of thermal stress, Vol. 24, pp 141-192.
Lindgren L. (2001), Finite element modeling and simulation of welding part 2: Improved
materials modeling, Journal of Thermal Stresses, Vol. 24, pp 195-231.
Lindgren L. (2001), Finite element modeling and simulation of welding part 3: Efficiency
and Integration, Journal of thermal stress, Vol. 24, 2001, pp 305-334.

Welding Processes 244
Tall L. (1991), Residual Stresses In Welded Plates--A Theoretical Study, Weld. J., Vol 43 (No.
1), P 10s-23s
Hibbit H.D. & Marcel P.V. (1972), A Numerical Thermomechanical Model For The Welding
And Subsequent Loading of A Fabricated Structure, Nsrdc Contract No. N00014-67-A-
019-0006, Tech. Rep. No. 2, Department Of The Navy.
Muraki T. et. al. (1975), Analysis Of Thermal Stresses And Metal Movement During
Welding-- Part I: Analytical Study; Part Ii: Comparison Of Experimental Data And
Analytical Results, J. Eng. Mater. Technol. (Trans. Asme), 1975, P 81-84, 85-91
Rybicki E.F. et al. (1978), A Finite-Element Model For Residual Stresses And Deflection In
Girth-Butt Welded Pipes, J. Pressure Vessel Technol. (Trans. Asme), Vol 100, P 256-262
Courant R. (1943), Variation Methods For The Solution Of Equilibrium And Vibration,
Bulletin Of The American Mathematical Society, 49. Pp 1-23.
Dong P. et al. (1998), Residual Stress Analysis Of A Multi-Pass Girth Weld, Asme Pvp-
Fatigue, Fracture, And Residual Stresses, 373, Pp. 421-431.
Feng Z. et al. (1996), A Finite Element Model For Residual Stress In Repair Welds, Asme Pvp-
Residual Stresses In Design, Fabrication, Assessment And Repair, 327, Pp. 119-125.
Karlsson L. et al. (1989), Residual Stresses And Deformatins In A Welded Thin-Walled Pipe,
Asme Pvp- Weld Residual Stress Andplastic Deformation, 173, Pp. 7-10.
Grong Q. & Myhr O. R. (1993), Modelling Of The Strength Distribution In The Heat Affected
Zone Of 6082-T6 Aluminum Weldments Mathematical Modelling Of Weld
Phenomena, The Institute Of Materials, London, Pp. 300-311.
Voss O. et. al. (1999), Consideration Of Microstructure Transformations In The Calculation
Of Residual Stresses And Distortion Of Larger Weldments, Mathematical Modeling Of
Weld Phenomena 4, The Institute Of Materials, Pp584-596.
Tenga T. L. & Linb C. (1998), Effect Of Welding Conditions On Residual Stresses Due To
Butt Welds, International Journal Of Pressure Vessels And Piping 75, 857864.
Li J. et. al. (2004), Studies On Characteristics Of Temperature Field During Gtaw With A
Trailing Heat Sink For Titanium Sheet, J. Of Mater. Proc.Tech. 147,328335.
Karlsson R. I. & Josefson B. L. (1990), Three-Dimensional Finite Element Analysis Of
Temperatures And Stresses In A Single-Pass Butt-Welded Pipe, Asme-Journal Of
Pressure Vessel Technology, 112, Pp. 76-84.
Junek L. et. al. (1999), Residual Stress Simulation Incorporating Weld HAZ Microstructure,
Asme Pvp- Fracture, Fatigue And Weld Residual Stress, 393, Pp. 179-192.
Vincent Y. et. al. (1999), On The Validation Of The Models Related To The Prevision Of The
Haz Behavior, Asme Pvp-Fracture, Fatigue And Weld Residual Stress, 393, Pp. 193-200.
Dubois D. et. al. (1984), Numerical Simulation Of A Welding Operation: Calculation Of
Residual Stresses And Hydrogen Diffusion, Asme Fifth International Conference On
Pressure Vessel Technology, Materials And Manufacturing, Ii, San Francisco, Ca, Pp. 1210 -
1238
Gatovskii K. M. (1980) & Karkhin V. A. (1980), Theory Of Welding Stresses And Deformations,
Leningrad Shipbuilding Institute, Rus.

Numerical Simulation of Residual Stress and Strain Behavior After Temperature Modification 245
Withers P. J. & Bhadeshia H. K. (20000), Residual Stress, Measurement Techniques, Mat. Sci.
Tech., 2000.
Borland D. W. (1994), Residual Stress Measurement Methods, Limitations And Significance,
Proc. 2nd Australian Int. Conf. Surface Engineering, Adelaide, March 1994, 114-121.
Noyan I. C. , Defining Residual Stresses In Thin Film Structures, Advances In X-Ray
Analysis, Vol.35, 461-473.
Guan Q. et al. (1986), Transient Welding Strain Measurement By Moire Technique.
Transactions Of Chinese Welding Society, 7(1)
Artem P.(2001). Ph.D thesis, computer simulation of residual stress and distortion of thick
plates in multi-electrode submerged arc welding.
Soul F. & Yanhua Z.(2005), Numerical Study Of Residual Stress Field During Arc Welding
With Trailing Heat Sink. Fluid Structure Interaction And Moving Boundary Problem,
Transaction Engineering Science,Vol.84,683-692, Witpress.
Soul F. & Yanhua Z. (2005), Thermo-Mechanical Analysis of Longitudinal Welded
Aluminum Alloy Cylinder Using 3D FEM, The Fifth International Forum on Aluminum
Ships, Tokyo.
Huang T. D. et. al. (2004), Fabrication And Engineering Technology For Lightweight Ship
Structures, Part 1: Distortions And Residual Stresses In Panel Fabrication , Journal of
Ship Production, 20(1), P.43-59,
Andersen L. F. (2000), Residual Stress And Deformations In Steel Structures - PhD, Technical
University Of Denmark, Department Of Naval Architecture And Offshore Engineering.
Conrady C. & Dull R. (1995), Prediction Techniques For Distortion Control, Report-Edison
Welding Institute.
Michaleris P. et al. (1999), Minimization Of Welding Residual Stress And Distortion In Large
Structures. Welding Journal, Welding Research Supplement, 361-366.
Michaleris P. & Sun X. (1997), Finite Element Analysis Of Thermal Tensioning Techniques
Mitigating Weld Buckling Distortion. Welding Journal,Vol.76, 451-457.
Conrardy C. & Dull R. (1997), Control Of Distortion In Thin Ship Panels. Journal Of Ship
Production, Vol.13, 83-92.
Guan Q. et a (1994) . Low Stress No-Distortion (Lsnd)Welding A New Technique For Thin
Materials. Iiw Doc. X-1190-89, Welding In The World, Vol.33.
Guan Q. et al. (1993), Dynamically Controlled Low Stress No-Distortion Technology And Its
Apparatus. Chinese Patent 9310690.8.
Guan Q. et al. (1996), Low Stress No Distortion Welding For Aerospace Structures. China
Welding, Vol.5,1-9.
Li J. et al (2004). Stress And Distortion Mitigation Technique For Welding Titanium Alloy
Thin Sheet. Sci.Technol. Weld. Join., Vol.9(5),451-45
Soul F. & Yanhua Z. (2006), Numerical Study On Stress Induced Cambering Distortion And
Its Mitigation In Welded Titanium Alloy Sheet. J. Sci. Technol. Weld. Join., Uk, Vol.11,
688- 693.

Welding Processes 246
Soul1 F. et. al. (2010), Residual Stress Correlation In Two Different Mitigation Techniques
Using FEA, Advanced Materials Research Vols. 83-86 (2010) Pp 1254-1261
Pavelic V. et al. (1969), Experimental And Computed Temperature Histories In Gas
Tungsten Arc Welding, Welding Journal Research Supplement, 48, 295s-305s.
Goldak J. et. al. (1984), A New Finite Element Model For Welding Heat Sources,
Metallurgical Transactions B, Vol. 15b, 1984, P 229-305.
E.M. van der Aa(2007), PhD thesis, Delft University of Technology
Chapter 11




2012 Mijajlovi and Mili, licensee InTech. This is an open access chapter distributed under the terms of
the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Analytical Model for Estimating the Amount of
Heat Generated During Friction Stir Welding:
Application on Plates Made of
Aluminium Alloy 2024 T351
Miroslav Mijajlovi and Dragan Mili
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/53563
1. Introduction

Two decades after its invention, friction stir welding (FSW) has the status of a novel and
promising welding technique that has not yet been fully described, investigated or utilized
in industry [1]. It is a process of joining two metals/alloys, with relatively simple equipment,
that initializes complex physical processes in/around the parts that are being joined,
resulting in the monolith structure of these parts.
FSW is a patented [2-4] solid state welding technique that needs no consumables or
shielding gasses (except in some special cases) for the creation of a welded joint. Welding is
performed with a specialized, usually cylindrical, welding tool which mounts into the
spindle of a machine that can rotate the tool around its axis (Fig. 1, a).
During the last few years, FSW has been standardized for the welding of aluminium [6, 7]
and its principle of operation has been fully described for single welding tool application
[7]. Welding parts (recognized as a base metal, workpieces) mount on the backing plate
(anvil) [7] and rigidly clamp in such a manner that abutting or dilatation of the base metal is
prevented. The rotating welding tool is inserted into the base metal by (axial) force at a
start point on the joint line and travels along it [7]. While it travels, the welding tool
machines material in the base metal in the zone near the traveling path and confines it
in the working zone in a mixture that is deposed beside the welding tool as a weld. When
the welding length is reached, the welding tool retracts from the base metal and welding
is completed.

Welding Processes 248

Figure 1. Schematic of FSW: a) principle of operation, b) welding tool, c) schematic of heat generation
during FSW [5]
2. Heat generation during FSW
During FSW, the welding tool (Fig. 1, b) slides over the base metal, stirring, deforming, and
mixing it. The base metal, anvil, and welding tool increase in temperature due the influence
of the welding tool on the base metal. This change in temperature is a sure sign of heat
generation caused by frictional contact that takes place during the welding process.
Thermodynamics recognizes several different types of heat transfer from a hotter to a colder
body [8, 9]. Both the heat transfer and heat as an energy type have been investigated for a
number of cases. However, a challenge appears when heat generation occurs as a result of
the contact of two bodies. Heat generation is a process of energy transformation that takes
place when one form of energy transforms into heat [8, 9]. This transformation is complex
and it depends on the nature of the contact between the bodies, delivered loads, what
materials are in contact, the surroundings, movement of the bodies etc. [9, 10]. Heat
generated during FSW is the product of the transformation of mechanical energy delivered
to the base metal by a welding tool. The transition of mechanical energy from the welding
tool to the base metal happens between these bodies [10, 11]. Understanding that heat
generates when a metallic body receives an energy boost and recognizing the dominant
physical processes involved in the contact between the welding tool and base metal (friction,
wear, adhesion, deformation, recrystallization of material, etc. [5, 11]), some might say that
heat during FSW is primarily generated due to friction and deformation processes that
appear during FSW [11]. Friction processes always appear in boundary layers and,
therefore, the frictional heat generates in the boundary heat generative layer. Deformational
heat appears wherever the deformation of base metal appears: in the boundary layer as well
as in zone of deformed material around the welding tool [5, 11-13]. Heat generates due to
other processes (e.g. infrared radiation, vibrations) but at a much lower intensity than
results from friction and deformation.
Mechanical energy primarily transforms into heat when the welding tool contacts the base
metal, while secondarily it transforms in deformed material around the welding tool (Fig. 1,
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 249
c). That is the reason why the heat generative layer is the primary heat generation source
while deformed material around the welding tool is the volume of material where
secondary heat generation sources appear [5]. Understanding the process of heat generation
and estimating the amount of heat generated during FSW are complex and challenging tasks
that requires a multidisciplinary approach. Estimating the amount of heat generated during
FSW aids in understanding the appearance of thermal stress and structural changes in base
metals inflicted by heat. Understanding the heat generation in FSW might help in selecting
the optimal technological parameters of FSW (e.g. rotation frequency, travel rate) from
aspects of minimal thermal stresses and deformations, energy consumption etc.
2.1. State of the art
Heat generation and heat transfer became a topic of research related to FSW during mid
1990s. However, understanding heat generation and heat transfer processes within FSW
requires understanding several other physical processes: material flow around the welding
tool, contact pressure inflicted by the welding tool, the friction coefficient, wear, change of
thermo-mechanical properties and heat transfer coefficients etc. Nandan et al. [14] gives a
review of thermal processes in FSW, from the invention of FSW until 2008.
Chao and Qi [15] have introduced a 3-D heat transfer model in FSW with constant heat
input. Constant heat flux at the shoulder of the welding tool, constant contact pressure and
pure Coulombs friction law for estimating shear stress, and heat were the main
assumptions of the model. The experimental welding of plates made of aluminum alloy
6061-T6 was performed and the temperature history of welding plates was estimated. Heat
input was adjusted (trial and error principle) until numerical and experimental
temperatures were matched. As such, this model is the first model developed for estimating
the amount of heat generated during FSW. Frigaard and Grong [16] presented a process
model for heat flow in FSW, where they assumed that heat is generated only by friction on
the tops of shoulders and probes. Heat input and friction coefficients were adjusted during
the welding process to keep the calculated temperature below the melting point of base
metal material. Heat input was a moving heat source with a linear distribution of heat flux
at the contact surface. Gould and Feng [17], and later Russell and Shercliff [18], have applied
the Rosenthal equation [19] for describing the moving heat source, heat flux distribution,
and heat transport within base metals, welding tools and the surrounding area. Models
consider friction heat only at the shoulder and use a finite difference method for a numerical
solution of the heat equation. Russell and Shercliff [18] based the heat generation on a
constant friction stress equal to the shear yield stress at elevated temperature, which is set to
5% of the yield stress at room temperature. The heat input is a pure point or line source.
Colegrove et al. [20] have used an advanced analytical estimation of the heat generation on
the welding tool with a threaded probe to estimate the heat generation distribution. The
results show that the fraction of heat generated by the probe is about 20% of the total
amount. Shercliff and Colegrove [21] developed a material flow model that investigates the
influence of threads on the probe on material flow. An advanced viscous material model is
introduced and the influence of different contact conditions prescribed as the boundary

Welding Processes 250
condition is analyzed. A thorough presentation of analytical estimates of the heat generation
in FSW and influence of material flow on heat generation is given, as well. Khandkar et al.
[22] introduced a torque based heat input model where experimentally estimated torque is a
heat source. Khandkar modeled advanced heat transfer within the FSW process with
frictional and deformational heat input into the process. Song and Kovaevi [23]
investigated the influence of the preheating period on the temperature fields in FSW. A
sliding condition of the welding tool over the base metal was assumed and an effective
friction coefficient and experimental plunge force are input into the heat source expression.
Schmidt and Hattel [12] have defined an analytical model for estimating the amount of heat
generated during FSW that recognizes the shoulder and the probe of the welding tool as
heat sources and concludes that about 89% of heat is generated at the shoulder. Heat has
friction and deformation components and the total heat is a sum of both with influence of
the contact state variable [12, 24]. The effective value of the friction coefficient was used in
calculations. Reliability of the previously proposed ideas and principles of heat generation
were summarized by Nandan et al. [25]. Nandan has performed FSW of dissimilar
aluminum alloys and his results have shown that a constant state variable (also referred as
an extent of slip) gives values close to sticking.
Further advances in heat generation and modeling included finite element analysis (FEA) in
FSW. Ulysse [26] presented a 3-D visco-plastic FEA model using the commercial software
FIDAP. The heat generation was determined to be a product of the effective stress and the
effective strain-rate. Results show that the model consistently over predicted the measured
temperatures probably from an inadequate representation of the constitutive behavior of the
material used in FSW. Steuwer et al. [27] used the experimentally observed mechanical
power as input in the model. The influence of tool loads on residual stresses was
investigated. Chao et al. [28] recognizes two boundary value problems in heat transfer: heat
transfer in the welding tool as steady state while considering heat transfer within
workpieces as transient. The amount of the heat that flows to the tool dictates the life of the
tool and the capability of the tool in the joining process. In discussions, it is shown that the
majority of the heat generated from the friction, about 95%, is transferred into the workpiece
and only 5% flows into the tool and the fraction of the rate of plastic work deformation,
dissipated as heat is about 80%. Heurtier et al. [29] presented a 3-D model based on the
fluid-velocity fields where the tool shoulder and the plastic strain of base material near the
welding tool were heat sources. The model has shown good agreement regarding the
numerical and experimental results. Santiago et al. [30] introduced a model with rigid and
visco-plastic materials in which the plates move towards the rotating tool and the material
flow at the interface is specified as a boundary condition. The results estimated from the
model correspond to the steady state of the FSW process that has been proposed by Chao
[28]. Schmidt [31] has adopted a fully coupled thermo-mechanical dynamic analysis model
also aiming to achieve the steady welding state in ABAQUS/Explicit.
Colligan [32] gave a conceptual model that describes dominant parameters affecting heat
generation including a detailed description of the existing literature and the principles of
specific physical processes in FSW, e.g. friction coefficient. Kalya et al. [33] investigated
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 251
torque, specific energy and temperature during FSW and gave a correlation between torque,
angular frequency and travel rate. The conclusion was a simple correlation between the
friction coefficient and mathematically modeled torque. Kumar et al. [34] proposed an
experimental setup for estimating the friction coefficient for different contact pressures,
temperatures and materials of workpieces. A number of results showed that the friction
coefficient varies during FSW from 0.1 up to 4. Setup and the mathematical model are
limitedly applicable due to the design of the setup, kinematical complexity of the FSW and
approximations involved in the mathematical model. Colligan [35] has investigated the
material flow around the welding tool using a tracer embed and stop action technique.
The results gave deep insight in the material flow patterns, influence of the thread on
material flow, influence of rotation direction, as well as some metallurgic and details about
heat transport in the zone of the deformed material. Nandan et al. [36] gave some
numerically estimated results about viscoplastic material flow, heat transport, viscosity
changing and material deposition behind the welding tool. Ouyang and Kovaevi [37] have
shown that material flow is always dissimilar along the joint line when welding either
similar or dissimilar materials. Material flow-patterns can be easily traced in the vortex-like
microstructure known as the welding nugget. Material in the zone of the nugget suffers
significant plastic deformation, thermal recrystallization and acts like a quasi heat source.
Lorrain [13] has explained in detail the shear layer of material and material flow patterns
around the unthreaded probe. The existence of the rotating layer in the material near the
welding tool is pointed out and it is concluded that material flow is significantly lower
when a welding tool without threads is used. Lack of material flow had no influence of the
strength of the weld.
2.2. Analytical model for estimating the amount of heat generated during FSW
Heat generation is an unavoidable process following the friction stir weld-creation process.
Since FSW is a welding procedure that uses a welding tool as an initiator of the joining
process of workpieces, the welding tool delivers activation energy [38, 39] to workpieces
and the joining of the workpieces is achieved while heat generates.
This study presents an analytical model for estimating the amount of heat generated during
FSW [5]. The model recognizes geometrical, kinematic, physical and energetic possibilities
of heat generation during FSW, recognizes dominant parameters affecting the heat
generation process and uses them to estimate how much mechanical power is transformed
into heat. Existing models for estimating the amount of heat generated during FSW [12, 15,
17, 20, 22, 23, 25, 26] recognize many parameters affecting the heat generation process. Some
of them are topology and geometry of the welding tool, technological parameters (tool
rotation speed n [rpm] or angular velocity [rad/s], travel rate vx [mm/s], tilt angle, etc.),
loading (axial force Fz, torque Mt etc.), physical phases of FSW, duration of the welding
procedure, duration of certain phases of the welding procedure, etc. Furthermore, these
parameters initiate other parameters that affect heat generation process: friction coefficient
, contact pressure p, shear stress , temperature T, mechanism of heat generation (defined

Welding Processes 252
over e.g. the constant state variable ), etc. However, presented models simplify FSW
assuming e.g. constant friction coefficient [12], constant contact pressure [15], pure frictional
heat generation [12, 15, 17, 23], heat generation only due work of the largest part of the
welding tool [20, 23, 25, 26], no heat generation when temperature in the workpiece reaches
melting point [22, 23] etc. Such assumptions are affecting the usability and the precision of
results derived by developed models.
The model presented here considers many of the previously analyzed parameters. Special
care in the model is given, beside estimating values of parameters, to the mutual
dependences between parameters and their influence on the heat generation process. Such
dependences are numerous and it is not possible to recognize all of them. Furthermore,
many of them are too difficult to be explained analytically and require the numeric
calculations and the experimental estimation/validation. These are the reasons why
analytical model considers only the most important dependences (Fig. 2, a).
Because of the nature of this approach, the proposed analytical model relies on three major
elements: analytic algebra, numerical calculations and experimental data [5]. The analytic
algebra is based on existing research and results but includes some improvements. The
algebra is developed for a complete welding tool, involves more dominant parameters in the
calculations than in previous models, recognizes more dependencies between parameters,
neglects fewer parameters and has a shorter calculation time. One of the improvements of
the algebra is the implementation of a numerical material flow model with respect to energy
balance in workpieces. The numerical calculations use adequate numerical procedures to
give good precision and convergence during a short-computing time.

Figure 2. a) Schematic of mutual dependencies between generated heat and dominant influencing
parameters [5, 44], b) Partial algorithm for generated heat estimation [5]
The analytical model gives precise results only if experimentally estimated parameters are
involved in the model. Furthermore, verification of the analytical model can be done by
comparing the results from analytical model with experimentally estimated results.
Experimental data is obtained during the welding of the workpieces made of Al alloy 2024
T351 and is used as input and for verification of the analytical model. Mutual dependences
of the parameters affecting the heat generation are derived in iterative work regime of the
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 253
analytical model: time and space are discretized and conditions numerically estimated for
the present discretized moment of time are the input for the future discretized moment of
time (Fig. 2, b).
Since welding tool is the main initiator of the welding and the heat generation process, it s
important to analyze the welding tool and its influence on workpieces as well as the
physical engagement of the welding tool while welding.
2.2.1. Active surfaces, active surfaces engagement and physical phases of the FSW process
A number of different types of welding tools have been introduced from 1992 until the
present [1, 40, 41]. They differ in shape, dimension, mechanical properties etc., and every
tool is applicable to a specific material and limitedly applicable to some others. However, all
welding tools have the same basics: they consist of at least one shoulder carrying at least one
probe directly involved in welding. Recently advanced bobbin tools [7] have appeared when
the welding tool has two shoulders. No matter how complex or simple the welding tool is, a
limited portion of the welding tool is in constant contact with the base metal and performs
the welding. The welding contact region (WCR) on the welding tool consists of three areas
called the active surfaces of the welding tool (ASWT). There are always three of them: probe
tip (pt), probe side (ps) and shoulder tip (st) (Fig. 1, b). Complete welding and all physical
processes following it appear on these surfaces or close to them [5]. Probe tip is usually the
smallest ASWT located at the top of the probe. It can be flat, curved or flanged, and rounded
at the corner where it connects with the probe side. The probe side is cylindrical or conned
ASWT sharing the same rotational axis with the probe tip. The area of the probe side is
enlarged by various threads or flanges that help in more intensive mixing of material into
the weld [41]. The root of the probe side connects with the shoulder. The shoulder tip is the
largest ASWT, usually flat or curved in a manner that creates a reservoir for flashed
material that come from the workpieces. It is confined to the top surface of the weld and has
a role in imperfection-free weld creation [11].
At the beginning of the FSW process, the welding tool is positioned above the workpieces
and the rotation axis of the welding tool is (nearly) perpendicular to the joint line.
After positioning, the welding tool starts to rotate at a constant rate (n revolutions per
minute, angular velocity of [rad/sec]) and slowly plunges into the workpieces in the
direction of the z-axis. That is the start of the welding process starts t=t0. Plunging stops
when the plunging depth is reached (t=t1, duration of the plunging is tpl=t1-t0). The plunging
depth is equal to the height of the workpieces or slightly smaller and it is achieved at a
constant travel rate of the welding tool of vz. The welding tool continues to rotate and dwells
until t=t2. During this time period (tdw1=t2-t1) the workpiece material is being prepared for the
welding: it heats and softens in the area near the welding tool. Afterwards the welding tool
begins a translational movement along the joint line (x-axis) at a constant travel rate of vx.
The rotation and translation of the welding tool make the workpiece material (near the
welding tool) deform, stick and mix into a monolith composition (weld) that is deposed in
the area behind the welding tool. Movement of the welding tool along the joint line lasts

Welding Processes 254
until the welding length l is reached, at t=t3. This period (tw=t3-t2) is the productive phase of
the welding process. Translation of the welding tool stops and the tool dwells at the end
point until t=t4 (tdw2=t4-t3). The welding tool then moves in z direction and leaves the weld
and workpieces. When the welding tool is completely removed from the workpieces (t=t5,
tpo=t5-t4) the welding process is over. The physical phases of FSW are shown in Fig. 3. In
certain circumstances, dwelling can be excluded from the welding process, however, a full
FSW process consist of the plunging phase (t0 to t1), first dwelling phase (t1 to t2), welding
phase (t2 to t3), second dwelling phase (t3 to t4), and pulling out phase (t4 to t5) [5].





Figure 3. Physical phases of the friction stir welding process
Active surfaces of the welding tool are involved differently in the welding process during a
complete cycle of welding, and engagement of every active surface varies during the cycle
(Fig. 4, b). The probe tip is involved in welding from the beginning of the welding process
until the end of the second dwelling phase (t0 to t4). Since the complete probe tip is fully
involved in welding, engagement of the probe tip is considered as maximal. The probe side
is involved in the welding process when intensive plunging appears (at tpl)[5, 42-46].
Engagement of the probe side rises with the rise of the probe plunge into workpieces. With
the end of the plunging phase (t1), engagement of the probe side reaches a certain value and
keeps it during the complete first dwelling phase. When the welding phase starts (at t2),
engagement of the probe side rises toward maximal. This value is reached when the
welding process stabilizes travel rate (s) vx reaches a steady value (at tpl) and it keeps
constant until the end of the second dwelling phase (t4) and afterwards it decreases. The
shoulder tip is involved in the welding process before the end of the plunging phase (tst) and
it reaches full engagement when plunging stops (t1) and keeps it until the end of the second
dwelling phase.
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 255

Figure 4. a) Welding tool used for experiments [5], b) Active surface engagement (ASE) [44]
2.2.2. Estimating the amount of heat generated during FSW
As previously mentioned, the heat generation process within FSW is a process that
transforms mechanical energy (power) into heat. If Q represents a heat transformation [5],
the total amount of heat generated during FSW - Qt is a function of mechanical power Pa
delivered to the welding tool:
[W], (0, 1)
t Q a Q
Q P (1)
The welding tool performs dual movement: translation (tr) and rotation (rot), and the total
amount of generated heat is the sum of translation Qttr and rotational-generated heat Qtrot:
+
t ttr trot ttr
Q Q Q Q
0
+
trot
Q (2)
The amount of translation heat is significantly smaller than amount of rotational heat [5, 12]
and it can be neglected in analysis.
Heat is generated at or near the ASWT [5, 10, 12] and the total amount of generated heat is
the sum of heat generated on every ASWT:
+
t pt ps st
Q Q Q Q (3)
where Qpt the amount of heat generated at probe tip, Qps the amount of heat generated at
probe side and Qst the amount of heat generated at shoulder tip.
Simplifying the analysis and assuming that the total amount of mechanical power
transforms into heat (Q =1), the total amount of heat becomes:

t tr a
Q Q P
(4)
Mechanical power depends on angular frequency and torque Mt and the total amount of
generated heat is:

Welding Processes 256

t t
Q M e = (5)
and
d d d d
t t t cont
Q M r F r A e e e t = = = (6)
where dFt - infinitesimal force, r - distance of a infinitesimal segment, dA - infinitesimal area,
tcont - contact shear stress in material.

Figure 5. Active surfaces of the FSW welding tool
Different topologies of active surfaces result in different amounts of heat generated on them
that give different expressions for estimating the amount of generated heat (Fig. 5). After the
integration of Eq. 6, the expressions for the analytical amount of heat generated on every
ASWT are, respectively:
d
3
2 / 2
2
0 0
2
d
3 2
d
pt cont cont
d
Q r r e t u et
t
| |
= = t
|
\ .
} }
(7)

2 2
2
0 0
1+tg d d 2 1+tan
2 2 2 2
h
ps cont cont
d d
Q z h
| |
e t u et
t
| | | | | | | |
= = t
| | | |
\ . \ . \ . \ .
} }
(8)

( ) ( )
3 3
2 / 2
2
0 / 2
2
1 tan d d 1 tan
3 2 2
D
st cont cont
d
D d
Q r r e t o u et o
t
(
| | | |
(
= + = t +
| |
(
\ . \ .

} }
(9)
where: d - nominal diameter of probe, D - diameter of shoulder, h - height of probe, o - cone
angle of shoulder, | - cone angle of probe.
There is heat generated by friction (frictional heat) and heat generated by deformation
(deformational heat) [5, 10-12]. Both types of heat appear simultaneously on every ASWT
and both influence one another. Considering both types of heat and their mutual influence
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 257
on one another, the total amount of heat generated on the probe tip, probe side, and
shoulder tip are, respectively:

( )
1
fr def
pt pt pt pt pt
Q Q Q o o = + (10)

( )
1
fr def
st st st st st
Q Q Q o o = + (11)

( )
1
fr def
ps ps ps ps ps
Q Q Q o o = + (12)
where heat indexed with fr represents frictional heat, heat indexed with def represents
deformational heat, opt, ops, ost a dimensionless contact state variable (extension of slip) at
the probe tip, probe side and shoulder tip, respectively.
The frictional and deformational amount of heat in equations 10, 11 and 12 for every ASWT,
using Equations 7, 8 and 9 with respect to the contact shear stress [12, 5] is:

cont
yield
p
t
t

, for frictional heat generation


, for deformational heat generation
(13)
where: -friction coefficient, p-contact pressure, tyield-shear yield strength of workpieces.
Beside geometrical dimensions of the welding tool (d, D, h, H, o, |, etc.) and technological
parameters of the process (e, vx), all other parameters (, p, tcont, tyield, T, Fz(t), Mt(t), opt, ops, ost,
t1, t2, tps, tst, etc.) necessary for the analytical model have to be estimated analytically,
numerically, experimentally or combining the estimation procedures.
Estimating the friction coefficient: Due to the complex kinematics of the FSW, it is difficult
to establish a straightforward procedure for estimating the friction coefficient in FSW.
Previous research recognizes the friction coefficient as a variable in FSW, but neglects the
variation and assumes a constant value throughout the complete cycle of FSW. Usually, the
friction coefficient within FSW, for a welding tool made of steel and workpieces of
aluminium is equal to 0.3-0.4 [12, 34].
Kumar et al. [34] proposed an experimental model for estimating the friction coefficient
during FSW. The model is based on the experimental estimation of the momentum of
friction and axial force, which are necessary for estimating the friction coefficient. Figure 6
gives the functional schematic of the measuring place for the estimation of the friction
coefficient. To estimate the coefficient of friction during FSW, it is necessary to estimate the
momentum of friction and axial force [5]. The momentum of friction is the multiplication of
the tangential force Ft(t) (measured at force sensor 10, Fig. 6) and length of the pole (friction
pole) Lt. If the diameter of the welding tools probe in contact is d(t) and axial force is Fz(t),
the friction coefficient can be estimated as [14, 34]:

2 0
3 ( )
,
( ) ( )
t t
z
F t L
t t t
F t d t
= > > (14)

Welding Processes 258


Figure 6. The measuring configuration for the momentum of friction and axial force: 1-welding tool, 2-
welding tools spindle, 3-shaft, 4-workpieces, 5-force sensor (axial force), 6-anvil, 7-backing plate, 8-ball
bearing, 9-fundamental bolts, 10-force sensor (tangential force), 11-pole
The proposed model gives approximate results and only for the first two phases of FSW
plunging and first dwelling. The model is not applicable to the welding phase because the
measuring system loses its stability when the welding tool travels along the join line and
the momentum of friction cannot be measured [5]. Without proper model for estimating
the friction coefficient during welding phase, friction coefficient has to be modeled.
Friction coefficient used for the analytical model is estimated regarding the experimental
results.
Estimating the contact pressure: Contact pressure p appears at the beginning of the
plunging phase as a result of axial load Fz(t) on the welding tool. Hertz [45] has proposed
the first model for distributing contact pressure if a cylinder with a flat base punches into
the plane, while Munisamy et al. [46] and Levytsky [47] have proposed models describing
contact pressure distribution and heat generation when the axis of the cylinder is tilted.
Distribution of the contact pressure p(r,t) delivered by the flat probe tip (Fig. 7, a) is [48]:



0
2 2
2
, , , 0
2
4
z
st
F t
d
p r t t t t r
d d r


(15)
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 259

Figure 7. Contact pressure distribution for a) flat and b) spherical probe tips [4, 42, 43]
If the probe tip has a spherical shape, contact pressure is distributed (Fig. 7, b) as [48]:
( )
( )
2
2 2
3
0
5
3
2
, 4 , , 0
2
z
st
F t E
d
p r t d r t t t r
d
= s < s s
t
(16)
where E represents the median modulus of elasticity estimated as:

2
2
1
1 1 wp
wt
wt wp
E E E
v
v

= +
(17)
and Ewt the modulus of elasticity of welding tools material, vwt the Poisons ratio of
welding tool material, Ewp the modulus of elasticity of the workpiecess material, vwp
Poisons ratio of the workpiecess material.
For engineering practice median contact pressure pm(t) gives good results:
( )
( )
( )
( )
0
2
4
, , , 0
2
z
m st
F t
d
p p t d t d t t t r
d t
~ = = s < s s
t
(18)
Research [5, 10, 42-44] has shown that contact pressure distributed to workpieces reaches
different values in different zones in some zones it overcomes the yield strength of
workpieces, while in other zones it has values lower than the yield strength (Fig. 8, c).
Existence of such zones multiplies the resistance of workpieces and plunging and intensive
plunging appears when [5, 10, 42-44]:
( ) ( ); 1.5-3
m eh yield eh
p t k T k o > = (19)
where: oyield(T) yield strength of workpieces in function dependence with temperature T.

Welding Processes 260

Figure 8. Contact pressure distribution [4, 42, 43]: a) probe side, b) shoulder tip, c) contact pressure
defining contact conditions [5, 10, 42]
Contact pressure delivered by the shoulder tip is similarly distributed (Fig. 8, b) with a flat
probe tip [5]. It appears smoothly since the shoulder tip is continuously involved in
welding. Superposed contact pressure delivered by the probe tip and shoulder tip [5] is:

( )
( )
1
1
2
1 4
,
4 ( )
, ( )
( )
,
st st
z
st m
d D
t t d t t t
F t
t t p p t d t
d t
D t t t

~ + s <

~ =

t

= s <

(20)
Contact pressure delivered by the probe side (Fig. 8, a) is a case of a modified cylinder in
cylinder contact problem [45, 49]. Threads on the probe side increase the complexity of the
analysis of the contact pressure distribution, however, with or without threads, median
contact pressure on the probe side is:

2 3
2 3
( )
,
( )
, 0,
x
m
F t
t t t
p p t
dh
t t t t

< < ~
~

s > ~

(21)
where: Fx(t) force in welding direction, h height of the probe/workpieces.
Estimating the tangential shear stress: When the deformation of workpieces appears, the
rotational layer of the softened material travels around the welding tool [5, 12, 35, 37]. This
is possible only if loads delivered by the welding tool inflict tangential stresses larger than
the shear yield strength. The boundary value of such tangential shear (contact) stress, from
von Misses yield criterion in uniaxial tension and pure shear [5, 12, 41, 42], is:
( ) ( , ) ( , ) / 3
cont cont yield yield
T T T t t t c o c = = = (22)
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 261
where: tcont(T) tangential contact stress in function of temperature T, tyield(T,c) shear yield
strength of workpieces material in function of temperature T and strain rate c, oyield(T,c)
yield strength of workpieces material in function of temperature T and strain rate c. Yield
strength of material is highly dependent on the temperature and strain rate, and the analysis
of tangential stresses within FSW requires the full temperature and strain history in of the
workpieces in a wide zone around the welding tool [5, 11, 14, 17, 23, 24, 27, 30]. However,
analysis of heat generation in FSW can neglect the influence of strain on the decrease of
yield strength and still maintain sufficient precision [12]. Neglecting is possible since the
maximal temperatures of the material reach about 80% [38] of the melting temperature
when strain has significant values due to near melting conditions in the material [18, 22].
Tangential contact shear stress is:
( ) / 3
cont yield
T t o = (23)
where: oyield(T) yield strength of workpieces material in function of temperature T.
Thermo-mechanical properties of Al 2024 T351 are given in [5, 12, 50-53].
Estimating the contact state variable: The contact state variable or extent of slip is a
parameter defining the influence of slipping in the heat generation process following the
difference in the velocity of the welding tool and material, and relates frictional vs.
deformational heat. It is obtained after curve fitting the experimental data regarding
measured velocities [12, 14, 54, 55]:

( )
( ) min min 0
0
1 1 ,
R
A
R
r
e A
R
e
o o o o
e
= + = (24)
where: omin minimal measured slip, o0 adjustable parameter depending on the material of
the workpieces, R maximal radius of the welding tool, e0 normalized angular frequency
of the welding tool (often the mid-point of the diapason of the measured angular
frequencies).
Early works [12] considered the extent of slip as a single value for a complete welding tool.
Experiments [5] have shown that the decomposition of the welding tool provides more
precise results for the extent of slip if estimated for every ASWT separately. For example,
when welding Al alloys with a steel welding tool with a threaded probe [12], with concern
for the ASE of ASWT, the partial extent of slip is:

( )( )
0 2 3 4
min min 0
2 3
0
0,
,
2
1 1 , ,
d
A
pt
pt pt d
t t t t t t
r
e A
t t t
d
o e
o o o
e

s < s <

=

+ =
s <

(25)

( )( )
1 2 3 4
min min 0
' 1 2 3
0
0,
,
2
1 1 , ,
,
d
A
ps
ps ps d
ps
t t t t t t
r
e A
t t t t t t
d
o e
o o o
e

s < s <

=

+ =
s < s <

(26)

Welding Processes 262

( )
( )
4
min min 0
4
0
0,
2
1 1 , ,
D
st
A
st
st st D
st
t t t
r
e A
t t t
D
o e
o o o
e

> >

=

+ =
s <

(27)
where: opt = 0.1, ops = 0.2, ost = 0.1, o0 =0-1 from [5].
Estimating the temperature history of workpieces: Estimation of workpiece temperature
requires knowing how much heat is generated during welding since heat influences
temperature increase and it has to be done in an iterative regime . An iterative regime
requires the discretization of time and space (Fig. 9, b), numeric calculations and significant
computing time [5]. Temperature history of workpieces and welding tool can be estimated
solving heat equations:
for workpieces)
2 2 2
2 2 2
(
w w w v
T T T T
c q
t
x y z

| |
c c c c
= + + + |
|
c
c c c
\ .
(28)

2
(for welding tool)
wt wt
wt wt wt v
T T T T
c r q
t r r r z z
r



| | | | | | c c c c c c c
= + + +
| | |
c c c c c c c
\ . \ . \ .
(29)
where: w density of the workpiece, cw specific heat capacity of the workpiece, w
thermal conductivity of the workpiece, wt density of the welding tool, cwt specific heat
capacity of the welding tool, wt thermal conductivity of the welding tool.

Figure 9. Workpieces, welding tool, bolts and anvil positioned for welding: a) realistic view with
dimensions and some technological parameters used in experiment [5], b) discretized view (primarily
and secondarily meshed with adaptive grid) [5]
Thermal energy generation source qv is directly affected by generated heat Qt and volume
receiving generated heat Vt:
/
v t t
q Q V = (30)
Initial conditions for such a system include the recognition of the initial temperature:
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 263

0 0 0
( , , , ) ( , , , ) T x y z t T r z t T = = (31)
Boundary conditions are complex due to the complex geometry of the welding tool and
complex kinematics. As an example, boundary conditions for the top surface of workpieces
involve convective and radiation heat transfer:

( ) ( )
4 4
0 , , 0 , , w i j k i j k
z h
T
T T T T
z
o oc
=
| | c
= +
|
c
\ .
(32)
where: o - heat transfer coefficient, o - Stefan-Boltzmann constant, c - thermal emissivity of
workpieces, T0 ambient temperature.
Boundary conditions on the contact between workpieces and anvil:

( )
aprox , , 0
0 0 0

w a w i j k
z z z
T T T
T T
z z z
o
= = =
| | | | | | c c c
= =
| | |
c c c
\ . \ . \ .
(33)
where: a the thermal conductivity of anvil, oaprox approximated heat transfer coefficient.
Boundary conditions between the welding tool and workpieces involve conduction between
the parts. Such condition is decomposed to a classical conductive boundary condition:

/ 2 / 2
w wt
r d r d
T T
r r

= =
| | | | c c
=
| |
c c
\ . \ .
(34)
and influence of heat transfer due to material flow [5]. The material of the workpieces in the
welding zone travels around the welding tool and partially carries its energy balance with it.
If analyzed in discretized space and during discretized time, nodes travel from one
discretized position to another and they carry its temperature, and while traveling they
get and lose a burst of heat. This model of material travel is based on research concerning
material flow around the welding tool [5, 13, 14, 20, 25, 26, 30, 34, 37], and applied in
numerical calculations of temperature and heat flow. The model is named node
substitution and replacement method - NSRM [5].
Results can be obtained analytically and numerically for temperature estimation, a finite
difference method, an explicit scheme with an adaptive grid were used, with the application
of algorithm NSRM. The numerical solution of Eqs. 29-30 with the application of Taylor
series for approximation of 2
nd
order derives and node positioning in discretized space is:

( )
1
, , " " " , ,
(for workpiece)
m m
i j k w x y z v i j k
w w
t
T K K K q T
c

+
A
(
= + + + +

(35)

" 1 '
, , " , ,
2
(for welding tool)
m m r
i j k wt z v i j k
wt wt i
i
K
K t
T K q T
c r
r

+
( | |
A
( | = + + + +
|
(
\ .
(36)

Welding Processes 264
where:






1, , , , , , 1, , , 1, , , , , , 1,
" "
2 2
1 1 1 1
1 1
, , 1 , , , , , , 1 , 1
" "
2
1 1
1
, , (a)
,
m m m m m m m m
i j k i j k i j k i j k i j k i j k i j k i j k
x y
i i i i i i i i
i i i i
m m m m
i j k i j k i j k i j k i j
z
i i i i
i i
T T T T T T T T
K K
x x x x y y y y
x x y y
T T T T T
K K
z z z z
z z



















1 1 1
2 2 2
, , , , , , 1,
2
1 1
1
1, , , , , , 1, ,
1
'
1 1 1 1
1, , , , , ,
"
1 1
, (b)
, , (c)
2
m m m m
k i j k i j k i j k
i i i i
i i
m m m m
i j k i j k i j k i j k
i i
r i i i i
i i i i i i i i
m m m
i j k i j k i j k
r
i i i i
T T T
T T T T
r r
K r r r r
r r r r r r r r
T T T
K
r r r r


















1
2
1, ,
1
2
1
, (d)
2
m
i j k
i i
i i
i i
T
r r
r r
r r

(37)

Figure 10. a) Discrete nodes with coordinates and temperatures, b) discretised space with a schematic
of node replacement and substitution method [5]
3. Experimental procedure applied on aluminium alloy 2024 T351
The analytical model for estimating the amount of heat generated during FSW [5] is closely
bound to the experimental research:
The analytical model gives realistic results only if some inputs into the model (axial force,
torque, momentum of friction etc.) are obtained during realistic welding;
Validation of the analytical model and verification of gained results is possible only if some
outputs from the model (e.g. numerically estimated temperature history of workpieces) are
compared with the experimentally obtained results.
Such demands of the analytical model require a workplace with measuring equipment.
Figure 11 gives a model of a realized workplace [5] where the welding of plates of Al 2024
T351 (some details given in Fig. 10, a) with the welding tool (given in Fig. 2, a) was
performed.
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 265

Figure 11. Workplace for FSW with measuring equipment: 1-workpiece, 2-welding tool, 3-anvil, 4-
welding tools spindle, 5-bolts, 6-backing plate, 7-force sensor, 8-torque sensor, 9-machines tool rest, 10-
bearing house, 11-clutch, 12-machines spindle, 13-fundamental bolts

Figure 12. a) Typical diagram of measured torque and axial force, b) Typical diagram of measured
torque, axial and tangential forces during FSW (plunging, first dwelling), c) Calculated friction
coefficient, d) Modeled friction coefficient [5]
Welding was performed on a universal lathe with a horizontal operational axis (z-axis).
Anvil, force sensor and backing plate are assembled and mounted on the machines tool
rest. Holes in the anvil (where fundamental bolts assemble force sensors) and the anvil and

Welding Processes 266
backing plate are larger than the diameter of the bolts and allow for the axial translation of
the force sensor and measurement of axial force. Workpieces (dimensions given in Fig. 9, a)
are bolted to the anvil and tilted over the machines tool rest at an angle of 1. The welding
tool and its spindle are clutched to the torque sensor that is mounted into machines spindle.
A rigid axial-radial bearing house between the clutch and the welding tool disables the
transmission of axial and radial forces from the welding tool to the torque sensor and
machines spindle. Such a design provides the correct measuring of axial force on the
welding tool. Second working/measuring configuration (given in Fig. 6) is used for
measuring the momentum of friction, axial, and tangential force at the welding tool in order
to estimate the experimental friction coefficient. The machines used for this measuring
configuration are vertical milling machines. The anvil and workpieces are mounted on an
axial bearing above the axial force sensor. The anvil carries a tangential friction pole
engaged in measuring the tangential force delivered by the torque of the spindle. As
previously mentioned, such a design of the measuring equipment provides usable results
only for plunging and the first dwelling phases. An infrared camera was used for measuring
both the configurations that were used to estimate the thermal history of workpieces and the
welding tool.

Figure 13. a) Schematic of sample extraction; b) Schematic of weld nuggets position; c) Tension sample
destroyed at boundary of weld nugget
The experimental procedure of welding had three stages: to get familiar with the welding
process, to get optimal technological parameters and to measure parameters necessary for
the analytical model while creating qualitative welds [5]. Figure 12, a represents a typical
diagram of torque and axial force measured during welding resulting in a qualitative weld.
Using a second measuring configuration, torque, axial and tangential force are obtained for
numerous conditions. Figure 12, b shows a typical diagram of these parameters, obtained
with optimal technological parameters. These values are used for estimating the
experimental value of the friction coefficient (Fig. 12, c). However, since the proposed
method gives limitedly usable values of the friction coefficient, based on experimental
results, the friction coefficient is modeled for usage in the analytical model (Fig. 12, d).
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 267

Figure 14. Hardness, joint (tensile) efficiency and observed dimensions of the welding nugget in
samples 1, 2, 3 extracted from welded workpieces
In order to investigate the properties of welds (tensile and bending efficiency, hardness,
metallic structure etc.), test samples were extracted from welded workpieces. Figure 13, a
gives a schematic of the sample dimensions and positions of samples in workpieces. All
welded joints used for sampling had a vortex-like structure of material called weld nugget,
located along the joint line (Fig. 13, b). All tested samples (1-3) from all tested welds have a
crack in the weld zone at the border of the nugget (Fig. 13, c).
Tested samples have shown a bending efficiency of about 12% (reaching a bending angle of
about 11 while samples from parent material have reached an angle of about 89). Results
of joint (tensile) efficiency, the samples hardness and the dimensions of the weld nugget in
test samples are shown in Fig. 14.
4. Results and discussion
Whereas the experimentally obtained results included in analytical model makes use of
some the necessary parameters affecting the heat generation process within FSW (contact
pressure, shear stress etc.) and the amount of generated heat is relatively easily estimated,
the numerical estimation of the temperature field of the wokpieces requires some
computational time [5]. Experimentally estimated temperature can be easily inputted into
analytical model in order to estimate the amount of heat generated during FSW and
computational time can be significantly shorten. However, temperature change is the main
product of heat generation and the verification of analytical model can be done via
temperature comparison [56].

Welding Processes 268

Figure 15. a) Effective heat transformation, b) Analytically estimated amount of generated heat [5, 56,
57]

Figure 16. a) Numerically estimated temperature T of welding plates (rendered image); b) Median
temperature of the material on contact with ASWT [5]
Figure 15, b gives the analytically estimated amount of generated heat during the welding of
plates made of Al 2024 T351 (the dimensions of plates and technological parameters of the
process are given in Fig. 9, a) with the welding tool with conned, threaded probe (given in
Fig. 3, a). There were six repetitions of the welding procedure and the same number of
applications of the analytical model. Characteristic moments of the process, operational
technological parameters, values of measured axial force and torque delivered to the
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 269
welding tool are given in Fig. 15, b, as well. The ratio of generated heat and engaged
mechanical power is effective heat transformation Qef (Fig. 15, a) and it varies between 60%-
100% in this application. The median value of effective heat transformation is 86.58% while
it reaches 90.25% during welding phase. The maximum generated heat is 2.9 kW, reached
during the plunging phase (when the shoulder tip is involved in the welding process at tst).

Figure 17. a) Temperature distribution in workpieces in plane normal to the joint line, b) temperature
history (experimental and numerical) of specific discrete nodes, c) location of selected discrete nodes
[56, 57]
Figure 16, a gives a rendered image of numerically estimated temperature in workpieces.
Rendering is done for the moment the temperature reached maximal value a few seconds
before the tool shoulder was involved in the welding process. Figure 17, a gives the
temperature distribution in a plane normal to the joint line for the same moment of time.
The temperature peak is slightly dislocated from joint line to the advancing side of the weld.
Analysis of heat generation required information about temperatures on contact between
ASWT and the workpieces. Figure 16, b gives the median temperatures of all ASWT during
the welding cycle. A comparison of the numerically and experimentally derived
temperatures is done for 24 discrete points on the workpieces with adequate points from the
numerical simulation (Fig. 17, c). Figure 17, b gives a diagram of both temperatures for two
selected points. The largest difference between temperatures was about 11% in point 1
(absolute of 12C) while other points had differences of 0.5-3.5% (absolute of 2-13C) [5].
If the engaged mechanical power and generated heat are compared, it is clear that they have
the same trend and notable changes in both are mostly connected with the characteristic
moments of time. Mechanical power consumption and heat generation are more intense in
the plunging and the first dwelling phases than in other phases. With a drop in the axial
force (at the beginning of the first dwelling phase), torque drops as well, which results in

Welding Processes 270
the stabilization of the heat generation process and a drop in the temperature of the
workpieces. The welding process is always in a quasi-equilibrium state: increase in
temperature lowers the generated heat, engages mechanical power and axial force;
decreases of the temperature, and increases power consumption and heat generation
(almost sinusoidal temperature character, Fig. 16, b). At the end of the first dwelling and the
beginning of the welding phase, the system (tool-workpieces-anvil) has reached nearly the
maximal observed temperatures. During the welding phase, the system constantly decreases
the temperature probably workpieces were overheated due to the long lasting first
dwelling phase (tdw127 s). Power consumption and heat generation drop as well while axial
force steadily rises until the moment when effective heat transformation reaches app. 85%
(t150 s, Fig. 15, b) and axial force reaches a value of 11-12 kN. Analyzed results from tensile
and hardness testing (Fig. 14), best distribution of hardness, smallest weld nugget and the
most efficient welds of app. 80% are reached in test samples no. 3 at the end of welding,
where heat transformation has a value near 85%. Samples 1 have the worst joint efficiency
(app. 55%), exhibiting W-shaped hardness distribution with great variation of hardness and
largest weld nugget. Welding in the zone of samples 1 is performed at highest temperatures
and with an effective heat transformation of app. 90% and the lowest axial force (7-9 kN).
5. Conclusions
The analytical model developed here utilizes analytical algebra, experimentally gathered
data and numerical calculations to estimate how much of the mechanical power delivered to
the welding tool is transformed into heat. The novel approach in the model focuses on the
recognition of the active surfaces involved in welding, engagement of active surfaces during
welding, recognition of dominant parameters involving heat generation and their
estimation, recognition of mechanisms of heat generation and their utilization, and
implementation of a new numerical model for defining the material flow around the
welding tool. Experiments and temperature based validation of the model are done on Al
2024 T351 5 mm thick plates.
Results from the model have shown that 60-100% of the mechanical power of the welding
tool transform into heat during FSW with a median of 86.58% in a complete welding cycle.
The median value of heat transformation during welding is about 90% which is in
agreement with previous results. Comparison of workpiece temperatures from numerical
calculations and experiments has shown a maximal relative error of 11% (about 13C) while
the maximal temperature of workpieces reached a maximal temperature of app. 394C (79%
of Al 2024 T351 melting temperature).
Heat generation appeared to be extreme in the welding-procedure-dependent process:
preheating of welding plates, derived by a long dwell of the welding tool at the beginning of
the welding, has significant influence on heat generation and the quality of the welded joint.
Too long dwelling overheats workpieces and the welding phase happens during
continuous transient cooling. In such conditions, heat transformation is near (and above)
90%, joints have 50% tensile efficiency and a large weld nugget. For conditions of heat
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 271
transformation of app. 85%, joints reach app. 80% of tensile efficiency and have the smallest
weld nugget. In both situations, it was necessary to have minimally 11kN of axial force to
acquire a qualitative weld. If the optimal technological parameters of FSW and the welding
tool are selected, heat management in FSW is of great importance for the quality of the weld.
Due to the low bending efficiency of investigated welding, alloy 2024 T351 is legally
considered to be a tough and a limitedly weldable alloy.
Author details
Miroslav Mijajlovi
*
and Dragan Mili
University of Ni, Faculty of Mechanical Engineering Ni, Serbia
6. References
[1] TWI, official web page: http://www.twi.co.uk/technologies/welding-coating-and-
fabrication/friction-stir-welding/ (01.12.2011.)
[2] Thomas W M et al., Friction stir butt-welding, Int. Patent App., PCT/GB92/02203, 1991.
[3] Thomas W M et al., Friction stir butt welding, GB Patent App., No 9125978.8, 1995.
[4] Thomas W M et al., Friction stir butt welding, UP Patent, No 5 460 317, 1995.
[5] Mijajlovi M. Investigation and Development of Analytical Model for Estimation of
Amount of Heat Generated During FSW (in Serbian), Ph. D. thesis, Faculty of
Mechanical Engineering Nis, University of Nis, Nis, Serbia, 2012.
[6] AWS D17.3/D17.3M:2010 Specification for FSW of Al Alloys for Aerospace Applications.
[7] ISO 25239: Friction stir welding Aluminium, parts: 1-5, 2011.
[8] Maxwell JC. Theory of Heat. Dover Publications, Inc. ISBN 0-486-41735-2, 1871
[9] Ili G. Radojkovi N. Stojanovi I. Thermodynamics II: Basics of heat distribution (in
Serbian), Vranje, Yugoslavia, 1996.
[10] Stamenkovi D. Djurdjanovi M. Tribology of the Press Fit Joints (in Serbian), Faculty of
Mechanical Engineering, Nis, Serbia, ISBN8680587486, 2005.
[11] Djurdjanovi M. Mijajlovi M. Mili D. Stamenkovi D. Heat Generation During
Friction Stir Welding Process. Tribology in Industry. 2009; 31(1-2):8-14.
[12] Schmidt H.J. Hattel J.W. An analytical model for the heat generation in friction stir
welding. Model. Simul. Mater. Sci. Eng. 2004; 12 (2004): 143157.
[13] Lorrain O. Favierb V. Zahrounic H. Lawrjaniec D. Understanding the material flow
path of friction stir welding process using unthreaded tools, Journal of Materials
Processing Technology, 2010; 210 (2010): 603609
[14] Nandan R. DebRoy T. Bhadeshia H.K.D.H. Recent advances in friction-stir welding
Process, weldment structure and properties. Progress in Mat. Sci. 2008; 53 (6): 9801023.
[15] Chao YJ. Qi X. Thermal and thermo-mechanical modeling of friction stir welding of
aluminum alloy 6061-T6, J. Mater. Proc. Mfg. Sci. 1998; 7 (1998): 215233.

*
Corresponding Author

Welding Processes 272
[16] Frigaard O. Grong O. Modeling of the heat flow phenomena in friction stir welding of
aluminum alloys, Paper presented at: INALCO 98. Proceedings of the Seventh
International Conference on Joints in Aluminum, Cambridge, 1517. April, 1998.
[17] Gould J.E. Feng Z. Heat flow model for friction stir welding of aluminum alloys,
Journal of Material Processing and Manufacturing Science. 1998; 7(1998): 185-194.
[18] Russell M.J. Shercliff H.R. Analytical modeling of microstructure development in
friction stir welding, Proceedings of the first International Symposium on Friction Stir
Welding, Thousand Oaks, CA June 1999.
[19] Rosenthal D. The theory of moving sources of heat and its application to metal
treatments [J]. Transaction ASME, 1946, 43(11): 849866.
[20] Colegrove P. Painter M. Graham D. Miller, T. 3 Dimensional Flow and Thermal
Modeling of the Friction Stir Welding Process, Proceedings of the Second International
Symposium on Friction Stir Welding, June 2628, 2000, Gothenburg, Sweden.
[21] Shercliff H.R. Colegrove P.A. Modelling of friction stir welding, Mathematical
Modelling of Weld Phenomen. Maney Publishing, London, UK, 2002; 6(2002): 927-974.
[22] Khandkar M.Z.H. Khan J.A. Reynolds A.P. Prediction of temperature distribution and
thermal history during friction stir welding: input torque based model. Science and
Technology of Welding & Joining. 2003; 8(3): 165-174.
[23] Song M. Kovaevi R. Thermal modeling of friction stir welding in a moving coordinate
and its validation, Int. J. Machine Tool Manufacturing. 2003; 43 (6): 605615.
[24] Hamilton C. Dymek S. Sommers A. A Thermal Model of Friction Stir Welding Applied
to Sc-Modified Al-Zn-Mg-Cu Alloy Extrusions. International Journal of Machine Tools
and Manufacture. 2008; 49 (2008): 230 238.
[25] Nandan, G.G. Roy, T.J. Lienert, T. Debroy, Three-dimensional heat and material flow
during friction stir welding of mild steel, Acta Materialia. 2007; 55 (2007): 883895.
[26] Ulysse, P. Three-dimensional Modelling of the Friction Stir-welding Process.
International Journal of Machine Tools & Manufacture. 2002; 42(2002): 1549-1557.
[27] Steuwer A. Peel M. Withers, P.J. Dickerson, T.L. Shi, Q. Shercliff, H.R. Measurement
and prediction of residual stresses in aluminium friction stir welds. Journal of Neutron
Research. 2003; 11(2003): 267-272.
[28] Chao Y. J. Qi X. Tang W. Heat transfer in friction stir welding-experimental, and
numerical studies. Transactions of the ASME. 2003; 125(2003): 138145.
[29] Heurtier P. et al. Mechanical and thermal modelling of Friction Stir Welding Journal of
Materials Processing Technology. 2006; 171 (2006): 348357
[30] Santiago D.H et al. Numerical Modeling of Welded Joints by the Friction Stir Welding
Process. Materials Research. 2004; 7(4): 569-574.
[31] Schmidt H. Hattel J. A Local Model for the Thermo-mechanical Conditions in Friction
Stir Welding. Modelling and Simulation in Mat. Science and Eng. 2005; 13(2005): 77-93.
[32] Colligan KJ. Mishra RS. A conceptual model for the process variables related to heat
generation in friction stir welding of aluminium. Scripta Mat. 2008; 58(2008): 327331.
[33] Kalya P. Krishnamurthy K. Mishra RS. Baumann J. Specific Energy and Temperature
Mechanistic Models for Friction Stir Processing of AlF35. Presented at: 2007 TMS
Annual Meeting Friction Stir Welding and Processing IV. Orlando, FL; 2007, pp. 113-126
Analytical Model for Estimating the Amount of Heat Generated
During Friction Stir Welding: Application on Plates Made of Aluminium Alloy 2024 T351 273
[34] Kumar K. Kalyan C. Kaias SV. Srivatsan TS. An Investigation of Friction During
Friction Stir Welding of Metallic Materials. Mat. and Man. Proc. 2009; 24(4): 438-445.
[35] Colligan K. Material Flow Behavior during Friction Stir Welding of Aluminum. Weld. J.
Suppl. 1999; 78(7): 229s237.
[36] Nandan R. et al. Improving reliability of heat transfer and materials flow calculations
during friction stir welding of dissimilar aluminum alloys. Welding Journal, 2007;
86(10): 313S-322S.
[37] Ouyang JH. Kovaevi R. Material Flow and Microstructure in the Friction Stir Butt
Welds of the Same and Dissimilar Aluminum Alloys. The Journal of Materials
Engineering and Performance. ASM International. 2002; 11(1): 51-63.
[38] Djurdjanovi M. Tribology, Welding, Friction Welding. [unpublished materials used in
studies at University of Nis, Faculty of Mechanical Engineering Nis, Serbia]. 2000-.
[39] Vukicevic M. Petrovic Z. Djuric S. Bjelic M. Oxy-fuel welding (in Serbian). Faculty of
Mechanical Engineering. Kraljevo. Serbia. 2007
[40] ivkovi A. Influence of Friction Stir Welding Tool Geometry on Properties of Welded
Joint of Alloys Al 2024 T351 (in Serbian) Ph. D. thesis. Faculty of Mechanical
Engineering University of Belgrade, Belgrade, Serbia 2011.
[41] Rai R. De R. Bhadeshia HKDH. DebRoy T. Review: friction stir welding tools. Science
and Technology of Welding and Joining. 2011; 16(4): 325-342.
[42] Mijajlovi M. Mili D. Anelkovi B. Vukievi M. Bjeli M. Mathematical Model for
Analytical Estimation of Generated Heat During Friction Stir Welding Part 1. Journal of
Balkan Tribological Association. 2011; 17(2) 179-191.
[43] Mijajlovi M. Mili D. Anelkovi B. Vukievi M. Bjeli M. Mathematical Model for
Analytical Estimation of Generated Heat During Friction Stir Welding Part 2. Journal of
Balkan Tribological Association. 2011; 17(3): 361-370
[44] Mijajlovi M. et al. Study About Friction Coefficient Estimation in Friction Stir Welding.
2011; Presented at: Balkantrib 11, The 7th International Conference on Tribology,
Proceedings, Thessaloniki, Greece, pp. 323-330.
[45] Hertz HR. On Contact Between Elastic Bodies(in German), Collected Works, 1(1885).
Leipzig. Germany. 1895.
[46] Munisamy RL. Hills DA. Nowell D. The solution of the contact between a tilted circular
rigid punch and an elastic half-space. Wear. 1995; 184(1995): 93-95.
[47] Levytsky VP. Interaction of Rigid Cylinder with Elastic Half-Space by Heat Generation
on The Contact Area. Int. J. Engng. Sci. 1994; 32(11): 1693-1702.
[48] Galin LA. Contact Problems; The legacy of L.A. Galin. Series: Solid Mechanics and Its
Applications (in Russian). Nauka. Moscow, Russia. 15(2008). 2008.
[49] Goryacheva I. Sadeghi F. Contact characteristics of a rolling/sliding cylinder and a
viscoelastic layer bonded to an elastic substrate. Wear. 1995; 184(1995): 125-132.
[50] EN 573-3:2007, Al and Al alloys Part 3: Chemical composition and form of products.
[51] EN 485-2:2008-10, Al and Al alloys Part 2: Mechanical properties.
[52] Metals Handbook: Properties and Selection: Non Ferrous Alloys and Special Purpose
Materials. 1992; 10(2). The American Society for Metals, ASM International, 1992

Welding Processes 274
[53] ASM Aerospace Specification Metals Inc. Aluminium 2024 T351 Data Sheet. c2012
[Cited 27 August 2012]. Available from http://asm.matweb.com.
[54] Deng Z. Lovell MR. Tagavi KA. Influence of material properties and forming velocity
on the interfacial slip characteristics of cross wedge rolling. Man. Science and Eng. 2001;
123(2001) :647653, 2001.
[55] Khandkar H. Zahedul M. Khan J. Thermal modeling of overlap friction stir welding for
Al-alloys, J. Mater. Process. Mfg Sci. 2001; 10 (2): 91106.
[56] Mili D. et al. Temperature Based Validation of the Analytical Model for Estimation of
the Amount of Heat Generated During FSW. Thermal Science. In press. 2012.
[57] Mijajlovi M. et al. Experimental Studies of the Parameters Affecting the Heat
Generation in the Friction Stir Welding Process. Thermal Science. In press. 2012.
Section 3




Sensing of Welding Processes



Chapter 12




2012 Ogawa, licensee InTech. This is an open access chapter distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Visual Analysis
of Welding Processes
Yoji Ogawa
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/53519
1. Introduction

Fusion welding process is very convenient method to join metal structures. It plays the most
important role on industrial production. The process itself is quite simple. The generated
heat sources such as arc and/or laser melt the limited parts of work pieces to be joined, and
the joint zone is unified after solidification of melting metal in the groove. However fusion
welding process contains much interesting research targets(Ogawa,2011). For example,
metal contains many elements those thermal properties are quite different. The surface is
oxidized. When the concentrated heat source strikes on the metal surface, some elements are
evaporating away from the original work piece. If the shielding is insufficient, some amount of
hydrogen and oxygen invades into the hot metal. Hydrogen reduces mechanical property of
the structure. And quick re-solidification produces quite different structure compared to
the original base metal which was qualified by thermal refining process. Usual fusion
welding process contains four phases such as solid, liquid, gas and plasma. Plasma is
quite hot, maximum temperature exceeds 10,000K. The temperature of metal in solid
phase to be joined is normally in room temperature. The temperature of molten pool in
the base metal is raised up more than melting point. But, this temperature is about 3,000K
or less. Then the temperature gap between liquid parts and plasma is tremendous. The
physical reactions in those regions include many unknown factors(Zacharia &
David,1993). These reactions occur in quite small area about one centimetre cubic space.
Therefor the moving speed of target in observing area is quite high. This is the main
reason why high-speed imaging technique is necessary on visual analyssis of fusion
welding processes. Almost all of welding processes are carried out in factories on the
earth. However, some welding processes must be carried out in deep water or in space.
These are another interesting area to study.

Welding Processes 278
2. Basics on imaging for fusion welding phenomena
2.1. General instruction for Gas Tungsten Arc Welding process
Basic concepts to study arc welding process had been constructed before 1970s(Pattee et
al.,1973). The most useful signals to understand arc welding process is arc voltage and
current. However, these two electrical signals are insufficient to understand actual process.
Then, high-speed filming technique was introduced. The most important matter to get
reasonable image of arc welding process is how to eliminate the meaningless light from the
process. Very strong light is radiated from the arc. This strong light protects to observe
actual welding procedure.
Figure 1 shows a fundamental concept of Gas Tungsten Arc welding (GTAW) process.
GTAW is one of the simplest of fusion welding process. Electrons emitted from tungsten
cathode impinge a base metal. Some parts of argon atoms, those are normal inactive
shielding gas to avoid oxidation of molten metal, ionize in an arc column. The arc column is
a channel of current between cathode and anode, and it includes the same number of ions
and electrons. The strong light is radiated by recombination reaction with ions and
electrons. Understanding of current density distribution is important to know thermal
transportation, and the distribution of this radiation suggests the temperature of arc column.
The GTAW process is materialized by the tight energy balance among the cathode, arc
column and weld pool. And there exist very complicated physical and chemical reactions in
each region.

Figure 1. Scheme of Gas Tungsten Arc Welding(GTAW) process.
Figure 1(b) shows the main typical regions of interests during GTAW process. There are
three major zones such as base metal, arc column and the tungsten electrode. The base metal
has three different regions. They are melting zone, solid zone and boundary between solid

Visual Analysis of Welding Processes 279
and liquid. The boundary zone is the most important on metallurgical point of view. Crystal
structure in boundary becomes large, and the mechanical property of the metal changes as a
result. Some defects like blowhole or undercut occur in this region. A melting area as called
weld pool is complicated. Electrons and atoms hit on the molten surface, and some atoms
invade into the pool. A surface of the base metal in front of the pool contains much oxide.
The melted oxide also invades on the pool. Many physical and chemical reactions occur on
and in the pool. Rapid liquid metal flow also occurs on and in the pool.
The tungsten electrode is expected as a non-consumable cathode. The cathode is mainly
heated by Joule heating by current passing through the electrode. The surface of the cathode
is heated by collision of atoms and ions from the outer space, and the top area, which called
cathode spot, is cooled by electron emission. Normal tungsten electrode contains a few per
cents of thorium oxide to improve emission ability. Thorium oxide on the cathode surface
evaporates during arc process. A cool surface of the electrode is normally oxidized. And
tungsten oxide is much easier to melt and evaporate compared to pure tungsten, then some
amount of tungsten oxide, 3 or more millimetres from the top, melt and move toward the
top where the arc generates. And the temperature near the top is hot enough to evaporate
the invading tungsten oxide. Evaporated tungsten oxide is easy to dissociate to tungsten
and oxygen. A temperature of middle region between the top and the area where tungsten
oxide evaporates is cool enough to crystallize the impinged tungsten atom from the outer
space. Very complicated physical and chemical reactions are occurred on the cathode
surface. A microscopic observation is required because normal diameter of the tungsten
electrode is 1.6-3.2mm.
The arc column existing between cathode and anode is hot enough where ionization,
dissociation, and recombination frequently occur. Particles in this region are heated and
forced by electro-magnetic field. Much power is conducted to the work piece by moving
electrons. Some power is lost to open air by radiation and conduction. Arc light is the
radiation loss of arc, and the frequency distribution of radiation has important information
of arc characteristics. It contains temperature distribution information in arc column. The
approximate power of arc radiation is very strong, and it protects precise observation of
cathode and anode surfaces. The energy transfer from the arc is crucial in the weld pool
formation. And the energy transfer due to convection becomes important in the weld pool
compared to heat conduction. Total thermal radiations from the weld pool and solid surface
are very weak compared to arc light. How to observe actual motion in and on the weld pool
under luminous arc column is the issue to study(Shaw,1975; Inoue,1981; Ogawa,2011) .
2.2. How to get good image
2.2.1. Spatial effect
Figure 2 shows basic principles to show spatial effect of arc light. Suitable selections of
exposure time, diaphragm and filtering set-up of camera are essential to get high quality
image. An external light is usually used to improve image quality of target which has a
strong light source in it. When the arc region is very small as shown in figure 2(a), normal

Welding Processes 280
light source is enough. When the arc region becomes large as shown in figure 2(b), stronger
light source is required to get clear image of the whole apparatus. Some halation occurs on
and near the arc. However, the arc region is also small enough to get clear image of
experimental set-up. When the arc region becomes larger as shown in figure 2(c), normal
light source is insufficient because arc radiation is quite high. The size of this figure is
normally used for observation of high speed imaging. We need some special technique to
improve image quality. Details of how to get high quality image of fusion welding process
are described in section 3.

Figure 2. Size effect on image quality.
2.2.2. Fundamentals of high speed imaging
High speed video camera is a digital processor with huge memory. Captured high speed
image data are processed as digital data inside of the camera. Depth of the captured data in
the camera is normally 10 to 14bits. However, data depth of output image/movie is normally
less than 8bits to fit normal picture/movie standards. Objects in three dimensional space are
captured onto two dimensional digital data. Many two dimensional data sets are stored time
by time as shown in figure 3(a). In case of monochrome camera, only one colour information
on each pixel is stored. In case of colour camera, it captures three kinds of colour
information on each pixel. They are red, green and blue. The arc light itself contains from
ultra violet to infrared regions. Normal colour camera is set to fit human eyes, so light which
is out of visible range of human eyes are avoided. But, ultra violet and infrared lights from
arc are so strong compared to normal scene, so it affects the colour tone of the captured
image(Ogawa,2008).
When the camera condition is set to get clear arc image, almost all back ground become
black as shown in figure 2(c). One simple solution to capture clear background image is
using a strong external light. Another good solution is using narrow banded interference
filter which protects arc light and pass the light from back ground. Spectroscopic data is also
useful to understand arc characteristics. Factors of interest to understand arc welding

Visual Analysis of Welding Processes 281
process are three dimensional space information, wavelength information and temporal
variations as shown in figure 3(b). Dynamic range of camera device or data depth of image
is also important factors for precise analysis.

Figure 3. Data structures of high speed imaging.
The image of object just on the focal plane is clear, and the brightness of captured data is
higher than those from out of focus position. When the objects is static, range information
can be detected by changing the camera position or focal distance as shown in figure 3(c).
Spectroscopic imaging is also important to understand the process behaviour as shown in
figure 3(d). If the object is stable and constant, static spectrum information can be detected
as shown in figure 3(e). Two dimensional digital data of each still images are stacked time
by time in case of high speed imaging. The data itself can be treated as voxel data as shown
in figure 3. Then statistical analysis becomes possible, and it becomes easy to choose
numerous points of view to analyse the whole process. This is the remarkable benefit of high
speed imaging to study on welding process.
2.2.3. Non-linearity on brightness
However, we need careful selection and understanding of observation conditions. Usual
camera is fabricated to fit human sense. Figure 4 shows effects of exposure time on data
brightness. The same static object is captured by different exposure time. One important fact
to realize is an effect of dark noise. Some special camera has special cooling system for
sensor device to avoid dark noise. However normal high speed camera has no special

Welding Processes 282
cooling system, then some dark noise is added on the data. And the values are affected by
temperature and exposure time. Another important fact is special fitting process on too
bright objects. There are no linearity in the highest range as shown in figure 5. Halation is
protected by this process, and the picture looks very clear. However, quantitative analysis
cannot guaranteed. The relation of actual brightness on object and captured data value
depends on camera. One typical relation is shown in figure 5. When we try to use high
speed camera for quantitative analysis like detecting temperatures, we need precise
correction of several different temperature object before actual observation.

Figure 4. Effect of exposure time on brightness distribution of data.

Figure 5. Typical relation between actual brightness of objects and data value.
2.3. Radiation effect
Surface conditions of a molten pool and a solid metal are quite different. The surface of the
molten pool is smooth like mirror, because of surface tension, and a light illuminated on the
molten pool reflects totally to the geometrical direction. On the contrary, solid surface has
rough surface, and the light illuminated on the solid surface is scattered to wide directions.

Visual Analysis of Welding Processes 283
This is an important difference for imaging of arc phenomena. Another important difference
is radiation efficiency by surface condition. Radiation from oxides is greater than that from
pure surface.
One of the major purposes of an imaging of arc is to provide a monochromatic two
dimensional image from a polychromatic object. Optical frequency distribution at
appropriate point has important information on arc temperatures at that point, such as
electron temperature, atom temperature and ion temperature. Normal colour picture is a
mapping of colour information from three dimensional scene on to a two dimensional
frame. In case of normal picture of arc column, one point of the picture contains integrated
intensity information in depth and integrated intensity on wavelength information. A single
colour picture which means a picture at appropriate wavelength, as called monochromatic
photography, and/or emission spectroscopy which measures information on wavelength
distribution at appropriate point are required to estimate the temperature at that point. They
are two dimensional data, and each point has information of intensity. The situations are also a
matter of time. Static or quasi-static phenomena is not usual for arc welding processes, they
usually changes time by time. So the time domain analysis is also required. Time domain
information is easily recorded as time series of two dimensional data as shown in figure 3.
The intensity of the data show integrated values across the depth of the space. The image
density measurements, as integrated intensity in depth, are input data for the Abel Integral
Equation. Radiance (at each wavelength) is obtained as a function of position. The local ratio
of radiances for two spectral lines then yields the temperature as a function of position.
Monochromatic imaging also has more direct value. Arc light is an exponential function of
radiation temperature. Total power of radiation is more than 1000 times of thermal radiation
from molten metal. This is the major difficulty to get clear image and weld pool
simultaneously. The highest value of radiation is estimated by Wien's law. The highest
radiation just near the boiling temperature of iron is in near infrared region between 950 and
1000nm. And radiation from arc at this region is low enough to get clear image of the molten
metal. Following formula are used to calculate temperature from image.

( )
2
5
/
34
2 1
1
6.6256 10 Planck's Constant
hc kT
hc
E
e
h Js

t

=
(1)

( )
( )
4
8 2 1 4
Stefan-Boltzmann's Law
5.67 10 deg Stefan-Boltzmann's Constant
E T
Jm s
o
o

=
=
(2)

( )
3
2.898 10 deg Wien's Transition Law
m
T m

= (3)

( ) ( )
( )
2
23 1
exp / Richardson's Theory
work function
1.387 10 Boltzmann's Constant
I A T e kT
e
k J K
|
|

=
=
=
(4)

Welding Processes 284
2.4. Spectroscopic and monochromatic imaging
Figure 6 shows examples of argon arc images on mild steel and SUS304 stainless steel. Left
hand-side pictures are over exposure images. Information of arc column is saturated, but
image of cathode is clear. Weld pool and base metal can be detected. In right hand-side
pictures, exposure time is too short to identify whole cathode surface, but plasma shape and
metal vapour from the weld pool can be recognized. The pictures on middle part are
suitable camera condition to recognize whole parts.
One good way to improve image quality is using circumferential filter of narrow band.
Figure 7 shows the effect of band-pass filter on arc images. Figure 7(a) shows normal colour
image, which reduce the intensity of all wavelength by ND(Neutral Density) filter. Images
of different wavelength are quite different, but all of each image are captured in the same
welding condition.

Figure 6. Example of argon arc image on metal plates.
The spectroscopic measurement of plasma condition is essential to understand plasma
temperature. Plasma temperature is a key element to understand energy and mass balances
in arc welding process. The problem for spectroscopic study on arc welding process is
existence of metal vapour from work piece and electrode. Ionization potential of metal is
much less than shielding gas. The metal vapour influences not only on ionization of plasma
but also an energy transfer in boundary regions between plasma and electrodes. A grating
monochromator is usually used to record spectra from representative arcs. Radiation
strength is a function of wavelength depends on particle temperature. Typical line spectra
due to transition of energy level indicate probabilistic number of temperature. In case of
argon shielded arc, suitable lines in the Ar I (neutral) and Ar II (singly ionized) spectra are
identified, measured and used to determine mean temperature of the species from the
Sahas equation, Boltzmann distribution and the radiation law, by the two line method. The

Visual Analysis of Welding Processes 285
ratio of typical two spectra indicates the temperature, and this method is a good way to
reduce measuring error caused by measuring device such as transmission loss of lens and
sensitivity of sensor device.

Figure 7. Example of single colour images.
The tendency is the same for thermometry. A monochromatic imaging is used to identify
the measuring point of spectroscopy, and it is also used to identify spatial distribution of
radiation. Spectroscopy by a prism is the simplest way to measure an interaction between
radiation and matter as a function of wavelength, but resolution is much less compared to
the grating monochromator. Figure 7 shows typical colour images, single colour images and
spectroscopic images for several shielding gases. A single colour image at 694nm and
spectroscopic image along the central line of the torch centre are captured by the same lens
simultaneously. Image through the lens is divided by a prism. One half of light is passing
through the band-pass filter of 694nm. The other half image is passing through the special
prism which is called Imspector, and it reaches to the sensing device. Monochrome sensing
devices are used for both cameras. Spectroscopic images are shown by pseudo colour
system to identify the quantitative differences. Helium has a few spectra, so it can be used to
identify the actual wavelength on horizontal position. Sensitivity of the image sensor is
affected by wavelength. Normally, it is the highest between 500nm to 700nm range. The
sensitivity over 900nm is one fourth of visible range or less. The sensitivity in the short
wavelength less than 400nm is also low. Reduction loss of lens is affected by the wavelength
also. Therefor precise quantitative comparison is difficult without correction of sensitivity
and resolution. However, qualitative consideration becomes easy by spectroscopic imaging.
Dynamic ranges of data depth and optical frequency range are the most important factors to
use measurement. A dynamic range in computer vision means data depth, and common
data depth is 8 bits. The data depth of 8 bits is completely small for optical diagnostic. It
usually requires more than 12 bits. Another important dynamic range is sensitivity of the
solid imaging devices such as CCD (Charge Coupled Device) and CMOS (Complementary

Welding Processes 286
Metal Oxide Semiconductor). Normal dynamic range of CCD camera is about 10
3
. This
value is much less for accurate measurements. A CMOS camera with specially tuned
electronic device for sensitivity as a wide dynamic range of 10
10-12
is available to use.
However data depth is still a major matter to protect scientific use. Practical methods to use
a solid devices for imaging are (1) using an well designed optical filter, (2) controlling the
shutter timing and duration, (3) using a wide dynamic sensor, and (4) correction of image
data those taken under different capture conditions as shutter speed, iris, and different
neutral density filters.

Figure 8. Effect of gas contents on arc plasma condition.
3. Observation of transient response
3.1. Observation of meta-stable stage of arc ignition
Argon and helium are usually used in GTAW process. Much smut covers on and around the
welding bead in case of helium arc welding. This is a quite simple evident of metal
vaporization during arc welding. Metal has much less ionization potential compared to
argon or helium. Therefore, effects of metal vapours on arc properties and subsequently on
weld bead configuration have been studied since 1950s(Pattee, 1973). Emission spectroscopy
and monochromatic imaging were used to determine the prominent metal species present
and distribution in arcs.
Temperatures of base metal and tungsten electrode are the same as room temperature until
welding process is starting. Temperatures of base metal and tungsten electrode are quickly
raised high. However they have some amount of heat capacity, so there are time rags to be
stable temperatures. High speed video is a good tool to capture the dynamic behaviour. But,
it is not so good to show movies to understand the dynamic behaviour. Capturing the video
finishes very short time. But watching the captured movie takes a quite long time. Showing

Visual Analysis of Welding Processes 287
the behaviour of dynamic response is another problem. Showing picture cue in time order is
one answer, but numbers of pictures to show and time step are difficult to decide. Using
streaking image on important location is good answer to understand the dynamic behaviour
at glance as shown in figure 9. Vertical line on the centre line of the electrode is drawn by
time. The top time chart shows the dynamic behaviour of centre line between arc start and
15seconds later. The behaviour at first 1 seconds changes much. It takes about 4-5 seconds to
be a stable condition. The second top picture shows the detail behaviour of transient state.
The third and the fourth pictures show the behaviour of the horizontal lines at cathode tip
and 0.5mm below the cathode tip. These pictures show the behaviour of plasma behaviour.
These streak images are good way to understand time response qualitatively. The pictures
shown on the right are single colour pictures at 694nm. They are taken simultaneously.
Taking pictures of different wavelength is a good way to consider actual behaviour of the
process. When the cathode becomes stable condition, upper range of the cathode becomes
brighter. Tungsten oxide on the surface becomes wet in this region, because temperature of
the tungsten electrode becomes high enough to melt tungsten oxide on and near the surface.
This high speed image shows from arc ignition to stable stage. However, video rate and
image resolution are insufficient to understand actual dynamic behaviour.

Figure 9. Typical example of GTAW at arc ignition stage.
3.2. Qualitative methods to show transient phenomena
Figure 10 shows examples of ultra-high speed pictures taken at 54k frames per seconds
(fps). A picture before No.1 is absolutely in black. There is a big change between No.1 and
No.2 pictures. There is brown coloured channel between the electrode and the base metal in
No.1 image. This brown colour, which envelops the electrode top, is brighter than the other
area. And three bright spots are shown on the base metal. These three spots have blue

Welding Processes 288
colour, which is a colour of recombination from metal vapour. One bright spot is remained
in No.2 picture. Blue colour zone looks like aurora occupies almost all space between the
electrode and the base metal. Brown coloured zone is disappeared. One bright spot appears
on the upper place of the electrode. There is a drastic change on arc behaviour at the ignition
moment. Using higher video rate is better to analyse in this arc ignition period, but this
video rate is the highest in this space resolution (320x256pixels) of this high speed camera.
Pictures among No.3 and No.12 look almost similar. This is another problem, when we
consider the process from captured high speed video. Capturing times of ultra-high speed
video is a few seconds, but it takes too much time to watch whole video. Much time to save
the data into a hard disk is another troublesome issue. One solution to recognize typical
transient behaviour is reconstructed the stacked images into streak image which contains
important information of time response.

Figure 10. Time series of high speed video at arc ignition stage.
Figure 11 shows an example how to reconstruct the movie to a still picture. Left picture is a
typical one shot at ignition stage. This is a typical case of arc ignition. Arc ignites in air, there
is no shielding gas. Cathode and anode are easily oxidized, and their reactions produce
much heat. Arc ignition in good shielding is shown in figure 12. Arc is very stable in this
case, but some drastic behaviour occurs in first 5 ms.
The video rate of 250fps as shown in figure 9 is insufficient to understand actual arc ignition,
because only two pictures in arc ignition stage can be captured in this video rate. Ultra high
video rate as shown in figure 11 and 12 indicates that quite complicated response is
appeared during first 5ms. High video rate is required for understanding of precise
transient behaviour, but transient states of welding process happens only short time range.
Welding process usually continues order of 10 to 100 minutes. And some unexpected

Visual Analysis of Welding Processes 289
transient behaviour during steady condition also occurs in short time. Therefore, high speed
video needs very huge memory size.

Figure 11. Example of time charts to recognize arc ignition stage in air.

Figure 12. Methods to show process behaviour GTAW in helium shield.
Using statistical data on time axis is another good way to understand spatial behaviour.
High speed camera captures many images in time series. Picking up the brightest value,
mean value and/or deviation for each pixel gives us good information. Figure 12(a) shows
the maximum value during arc ignition stage of 5ms. A whole time chart in this arc start
duration along 15ms is shown in figure 12(d). The picture of maximum values gives us
typical quantitative information on spatial behaviour. A locus of spatters indicates particle
size, velocity and flying direction with its origin. Sizes and positions of anode and cathode

Welding Processes 290
are apparently appeared as bright zone. A picture of mean value is almost the same as
normal still picture. An image of standard deviation indicates the active regions. The image
of maximum values emphasizes singular situation like sputtering. So, this image is useful to
identify the place somewhat abnormal situation exists. On the other hand, image of
standard deviation usually hides one time irregular event.

Figure 13. Examples of drawing methods to show time response.
Streak image is convenient for quick understanding of time response. And there are many
points of view to get valuable information on time response as shown in figure 13. One is to
check the typical line like centre line of the cathode as shown in figure 13(d). The arc process
usually assumes as cylindrical symmetry, however this assumption is always wrong during
arc ignition stage and improper welding condition. These conditions are main cases when
we need the analysis by high speed camera. However, difference between the streak image
of max value and it on the centre line suggests much good information for analysis. A streak
image of mean value is also important to recognize global time response of the process.
Vertical streak image is simple compared from the horizontal streak image. There are typical
three different area such as the cathode region, plasma region and the anode region on the
horizontal streak image. They are shown in figure 13(d', g, h and i), respectively. Pictures on
typical moments are essentially important to understand the process. We can reconstruct the
three dimensional behaviour of the process to use typical features of streak images in our
brains. However, we need some other pictures for reconstruction of spatial features.

Visual Analysis of Welding Processes 291

Figure 14. Examples of methods to show typical characters.
Figure 14 shows a synthotic image assembled different feature images. Figure 14(b) shows
mean values of stacked images in blue colour. Figure 14(c) stresses one time feature in red,
used values are maximum values minus mean value. Figure 14(d) shows deviation in green
colour. Figure 14(a) is a gathered image among these three colours of different features. This
kind of synthotic image which contains different feature helps to understand the spatial
property at any periods. Discussions of this paragraph are somewhat sensitive or
qualitative. Merit of using high speed camera is that captured huge data is digital, and
quantitative analysis is expected in this field.
Figure 15 shows time response of unsuitable welding condition. Behaviour of arc ignition
stage is almost the same as suitable condition. Because high frequency power source assists
arc ignition. But current is too low to heat up the cathode temperature to keep steady arc.
When the high frequency power is shut off, cathode spot (arc ignition area) becomes to
move irregularly. It becomes stable from after 243ms of arc ignition. However, arc spot
exists on the shoulder of the cathode. A single colour image of 957nm is captured in this
case to get in good contrast to show arc and cathode spot. Reactions on cathode and anode
are high-lighted in green, and reaction in plasma is high-lighted in red. Colour tone suggests
the location of happening.
Colour image contains much information. But, colours captured by cameras have different
characteristics. Captured image by usual colour camera is set to fit human sensitivity on
natural scene. Radiation from arc has discrete line spectra. Therefore colour tone of captured
arc image is very different by the makers and sensor kinds. We usually use (R,G,B) colour
sets on pictures. Some cameras use (Y,U,V) colour system to proceed data in the camera.
YUV colour system is reasonable on natural scene which has continuous property on colour
frequency. Colour tones on arc welding process are different by camera kinds. Some good
cameras have ability to correct colour tones, but it is very difficult to make the same colour
tone from different cameras.

Welding Processes 292

Figure 15. Examples of time response on not suitable cathode geometry.
3.3. Quantitative methods to show transient phenomena
Brightness is only apparent quantitative data at the first stage of analysis. Many quantitative
elements such as cathode size and brightness distribution of cathode spot, maximum
brightness of arc, arc area size, brightness distribution in arc, pool size, mean brightness of
pool, metal flow speed on pool, etc.. The brightness is a good indicator to pick up unusual
feature. One problem for analysis of captured data is a data depth. Normal data depth is
8bit, this means that digital range of brightness value is from 0 to 255. This data range is so
small for actual arc welding process. When we focus the analysis in arc ignition period,
brightness value of major target on this stage is less than those in steady state.

Figure 16. Time response of brightness in each frames.
Figure 16 shows the example of brightness values. This figure show maximum value(top),
50th value from the top, 1000th value from the top and mean values for each colour. Camera
condition is set to get clear image of sputters during arc igniton period, so rather over
exposure condition is used, and some data in bright area are saturated. Top 1000th data is
appeared in the graph, and many of top 50th data also appeared in the figure. Data size of

Visual Analysis of Welding Processes 293
this video is 320x256(=81,920elements), then top 1000th data means top 3% value. In this
meaning, capturing condition of this figure is almost the best to understand on arc ignition
stage. Figure 16 also shows brightness on the end stage. Brightness increases with time until
being on the steady state. So, all top values are saturated. Radiation of plasma suddenly
stops with arc expiration, however many hot particles remains in space. Radiation from hot
metal vapour is only appeared on this ending stage as shown in figure 12(h), so mean value
of blue is increased at this ending moment.
Mean value is low compared to top values, because it contains low values in dark back
ground. Figure 17 shows the difference of mean level for total image and it on arc and
relating area. Tendency of both data is almost the same except sensitivity on alteration. A
power source uses inverter control at high frequency. Brightness of arc changes with this
frequency, and the frequency is close to image capturing rate of 54kfps. So some interference
occurs as beat plotted in figure 17(b). However area size of target is almost the same, so mean
value for whole image has some sense. Spectroscopic high speed imaging is also carried out,
but video rate is 2kfps. Time charts of horizontal distribution shows the difference. Radiation
by recombination of atom is essential at arc ignition duration until ion produces.

Figure 17. Examples of brightness characteristics.
Estimation of temperature on cathode and anode is important to understand welding
process. Using thermal radiation value is good way to estimate temperature distribution
even it usually over estimates the temperature by arc influence. Another way to estimate
cathode temperature is using cathode spot area size. Electron density is a function of
temperature, so estimation of mean temperature of cathode spot becomes possible when we
count this area size from image. Figure 18 shows effect of gas contents on cathode

Welding Processes 294
temperature. Cathode spot in argon is concentrated on tip. Spot area size increases with
helium additionCathode spot size suddenly increases when helium content exceeds 25%.
Cathode temperature becomes high with cathode size enlargement. This sudden change
causes by cooling effect of electrode. When the cathode spot is small and it locates on the
top, current passing through the cone shaped electrode is heated efficiently by concentrated
current in the cone shaped electrode. Cooling effect acts also efficiently by conduction in the
cone. When cathode spot is enlarged by helium addition in argon, heating efficiency
decreases. Heating by hot helium collision onto the upper position of the electrode also
increases.

Figure 18. Effect of gas contents on arc behaviour.

Figure 19. Effect of pressure on arc behaviour and electrode temperature.
Temperature change by ambient pressure is also estimated as shown in figure 19. Shielding
gas is pure argon in this case, but argon is fulfilled in an experimental pressure chamber in
this case. So, there is no flow of argon along the electrode, cooling action by shielding gas is
not existed. And cooling system of cathode is different, those are reactions that cathode
temperature at atmospheric condition is higher than the temperature shown in figure 18.
Argon arc in low pressure looks like helium arc at atmospheric condition.
Figure 20 shows dynamic response of thermal radiation. Both electrodes are tungsten.
Upper one is cathode, and lower one is anode. This movie is captured through interference
filter of 532nm. Pseudo colour display is chosen to show brightness difference, because data
depth of this high speed camera is 10bit. Much heat is lost by emission of electron at

Visual Analysis of Welding Processes 295
cathode. Anode is heated by impingement of electrons. This is the reason why anode is
much brighter than cathode. Figure 21 shows effects of wavelength on brightness. Upper
pictures show original monochrome image, and lower pictures show pseudo colour image.
Thermal radiation at short wavelength is much less compared to long wavelength, because
the highest radiation occurs at about 950-1100nm. Capturing conditions are set in proper
value for each wavelength. Brightness is also normalized, so contour line shows
approximate brightness distribution.

Figure 20. Dynamic response of thermal radiation.

Figure 21. Effect of wavelength on radiation.
Estimation of temperature from brightness is simple method. But correction of obtained data
is difficult. There are many unknown factors to correct. One simple way to correct the data
is using the data at solidification area. Latent heat of solidification causes some typical
feature around this area. When the brightness data is arranged as time chart, the same value
continues at solidification stage as shown among point b and c in figure 22. Period between
a and b is melting stage without arc influence. Dropped values from arc stage to no arc
stage are about 1500 at point D, 1000 at point C and 750 at point A. These values are affected
by radiation from arc. Point A is not melted. Solidification starts at b, and it ends at c when

Welding Processes 296
almost all metal near the point c is solidified. Data during melting period, which is between
a and b, are almost the same. Brightness data increases when solidification starts, surface
of solid state is rough and it is covered by oxide, so radiation efficiency is high than the
liquid state. One problem is why brightness data at C and D are different. Time period
during solidification is longer at Point D, this is reasonable because heat capacity on fat area
is higher.

Figure 22. Measurement of surface temperature.

Figure 23. Example of single colour video for arc ignition stage.
Figure 23 shows temperature distribution calculated by brightness data. Argon is used as
shielding gas in this case, and bead on plate welding is carried out. Used band-pass filters
are 957nm and 970nm. Capturing condition is set to fit the brightness on tungsten electrode
becomes just below the saturation. The reason why near wavelength is used is to estimate
the influence from arc radiation. Calculated results for both cases are almost the same.
Temperature at top position becomes high at early stage of arc ignition. The arc is
concentrated at the top, therefor the temperature becomes high. Temperature becomes
stable about 2 to 3 seconds later. Values of temperatures are higher compared to those as
shown in figure18. Reasonable correction of brightness data is necessary.

Visual Analysis of Welding Processes 297
Figure 24 shows temperature distribution of bottom surface of the welding pool. There is no
affection by arc radiation, and boundary between liquid and solid is clearly recognized in
this case. Melting temperature can be used to correct brightness data to temperature.
Brightness on solid area is higher than that on solid area, therefore different fitting formula
are used to determine the temperature value. Inside of red colour region is molten pool, and
precise temperature distribution on welding pool is drawn in right. Upper picture shows
early stage of welding, and lower one is the distribution in steady state. Length of welding
pool becomes longer.

Figure 24. Pseudo colour image of temperature distribution.
4. Analysis of steady state
4.1. Effect of active flux on arc behaviour
Figure25 shows effect of active flux on behaviour of welding process. Upper pictures are
captured on slant position. Lower pictures are capture in horizontal position to watch metal
vapor on the pool. Major difference is size of anode area. Anode area for normal welding is
wide. Anode area for active flux is narrow, and some vapor jet is shown on anode area. Next
difference is position of metallic vapour color on the cathode. Metal vapour in arc decreases
plasma temperature, because ionization potential of metal is much lower than argon and
helium. And metal vapour is fully ionized in arc. Metal ion moves to cathode by electric
field. Anode area for active flux is very narrow. This means that electron is constricted to
this size, and almost all electrons impinge on this area. On the contrary, electron for normal
case is wide spread to broad anode area.
Figure 26 shows the dynamic response of arc behaviour from normal area to an area where
active flux is painted. Arc starts from left(normal area) to right(active flux area). Anode area
on normal case is wide. When an welding pool reaches to active flux area, melted active flux
invades on to the pool. And anode area is pushed to rear side by invaded flux layer. Upper
pictures and lower pictures were captured by different makers camera. A camera captured

Welding Processes 298
images in figure 26 is also made by different maker. Cone angle of electrodes, and surface
treatment, and captured date are, but welding condition and material are the same. Tones of
colour are quite different for these pictures. Another difference between normal arc and
active flux is pool behaviour. Front position of the pool becomes closer on active flux case.
And there is no change on pool length, so pool end moves to rear on active flux. Vibration
level increases on active flux, many small ripples are produced.

Figure 25. Effect of active flux on arc process.

Figure 26. Effect of active flux on arc and pool behaviour.

Figure 27. Effect of active flux on arc plasma.

Visual Analysis of Welding Processes 299
Figure 27 shows typical pictures of plasma configuration and spectrum distribution in
pseudo colour. Arc is generated in argon. Left pictures are single colour images at 950nm,
which range has no strong spectra for argon, for normal stainless steel and stainless steel
with active flux. These pictures are displayed in pseudo colour to intense the difference of
both conditions. These are single colour images, and brighter point indicates higher
temperature. Upper right picture shows spectrum distribution along central line of cathode
for normal welding. Lower right picture shows the difference between normal case and
active flux. The region where the brightness for normal plate is higher than it for active flux,
displayed in green colour. Red colour shows the opposite case. The intensity shows amount
of the difference of brightness. Radiation from normal plate near the molten pool is larger.
On the contrary, the brightness for active flux is higher in the electrode surface and outer
space of main plasma as shown in red colour. Radiation from active flux area near surface is
very low.
Actual physical and chemical process acts on brightness. Brightness on cathode is low. It is
difficult to recognize the difference to watch normal video by human eye. However there is
some difference of frequency on space and/or time. Pseudo colour display is good method
to show spatial difference as shown in figure 28(a,b). Small difference due to chemical
reaction also can be extracted as shown in figure 28(c,d). Melted thorium oxide on electrode
moves from upper side to top position. Behaviour of this chemical and physical reaction
becomes visible by some numerical treatment. These reactions remain the evidence on the
electrode. These evidence can be watched by SEM and EDM.

Figure 28. Effect of oxygen for reactions on cathode.
4.2. Effect of gravity on arc behaviour
Figure 29 shows effects of gravity on welding process. High speed video is captured by drop
tower experiment. Height of free drop zone is 10m, and time duration of micro gravity is
1.3sec. This time period is short, but it is enough to detect transient motion from normal
gravity condition to micro gravity condition. There is no time limit to continue welding
process before drop trial. Shape of molten metal is clearly affected by gravity. There is no

Welding Processes 300
apparent change on molten metal flow on the weld pool. However, inertial force acts on
metal flow, and the time response of inertial force is unknown. Reconstruction period to
balance static forces is finished in 10ms, and some vibration by overshoot motion remains
around 10ms.

Figure 29. Results of drop tower experiment.
4.3. High speed imaging of metal transfer on GMAW process
High speed imaging of wire melting and droplet transfer phenomena have been carried out
for long time. Silhouette imaging by using of strong external light was essential for observing
metal transfer during Gas Metal Arc Welding (GMAW) process. GMAW process has some
periodical vibration on metal transfer. Using band pass filters in near infrared wavelength
becomes convenient because it contains information on temperature(Ogawa, 2004).
Figure 30 shows the typical metal transfer of GMA welding by interference filter of 950 nm
without external lighting. Streak imaging is also useful to show dynamic behaviour. The
reactions at the plasma/metal interface include oxygen removal at the anode and the
discharge and pick-up of oxygen ions at the cathode. Figure 30 shows the reaction of oxygen
invaded into the molten metal in the melted wire. A combination of invaded oxygen and
carbon in wire makes carbon oxide gas. The gas is abruptly expanded by high temperature
and exploded on the way to the work piece. Streak image shows points for spattering and
unusual situation. Spatters are recognised as spike lines to the outsides from the wire centre,
and the flying speed is recognised as its locus angle. Abrupt expansion of the droplet is
recognised as an irregular knot. This scene is automatically detected by image processing of
this figure. The merit of the high speed imaging system is that it uses digital data; therefore,
effective analysis can proceed automatically(Ogawa et al., 2003).
A single colour video is a kind of thermal image, and it is presented in pseudo colour to
emphasise the physical changes. Cathode spot exists in the first frame. The melting part at
the top of the wire is growing, and the region between this melting part and the solid wire
becomes slender. Current density of this slender portion becomes high, and temperature of
this region increases quickly. Pinch force is also acted in this region. The combination of
these forces drives to release the metal. Arc is soon generated between new tip and droplet

Visual Analysis of Welding Processes 301
with some small spatters. This set of images gives a visual representation of transfer modes
in GMAW.

Figure 30. Time chart to pick up abnormal condition.
4.4. High speed imaging of Laser Welding process
Laser welding is very high speed welding method. Hybrid system such as Laser Arc hybrid
and Laser Hot wire hybrid are often used to improve joining efficiency of large structures.
Normal observation of arc welding process uses fixed torch system. Work piece moves
during welding. High speed camera is heavy and big to carry on laser torch. When moving
object is captured by fixed camera, object to watch moves in the scene as shown in figure
31(a). Reconstructing the scene as static torch system is easy as shown in figure 31(b).
However, this system needs wide range of image size. Static coordinate observation system
saves image size and/or improve spatial resolution as shown in figure 31(c).
Torch moving system has good points to present information on quality of whole welding
result as shown in figure 32. These pictures are produced by using of histogram
information. Lights have several infomations of their origin, and they appear in statistical
features. Behaviours of arc, fume and sputters are apparently drawn by statistical features.
Laser itself is invisible. But laser acts on fume and plume, and some statistical change
appeares on its value. So laser channel can be shown from reconstructed image as shown in

Welding Processes 302
figure 33. Figure 33 shows pseudo color display of mean value image, original mean image
and deviation images, respectively. Laser beam channel is apparent in this figure.

Figure 31. Comparison of coordinate system.

Figure 32. Exsample of imaging methods..

Figure 33. Statistical image of laser arc hybrid welding.

Visual Analysis of Welding Processes 303
4.5. Monitoring and evaluation of GTAW process
Figure 34 shows a training system for weldrs of one side butt welding of thin stainless steel.
This system uses four cameras to identify welder's skill. Welding trials are carried out by
manual operation. A torch camera captures weld pool and arc. Camera 2 and 3 are fixed on
carriage to capture surface and bottom situations of welded plate. Camera 4 is a fixed
camera to observe welders motion. The camera 3 is the most important camera to identify
the weld quality. This camera captures the conditon of bottom pool, and the system
indicates analysed status of penetrated condition by sound on real time. Five tones are used
to notice the actual condition to the welder. The welder can watch bottom situation by small
LCD monitor inside the cover face. This monitor indicates only the image from camera 3.
Voltage and current signals are also recorded and shown on the screen with colour. When
the signal is out of suitable range, normal green colour changes to red. The collected data is
stored in the data folder. In the same moment, a mentor of the welder tells to the welder
about important points, and this voice is also recorded in the system. The welder can watch
his own operation to select his data. Almost the same features are reproduced by the system
with mentors indicated voice. Several reference information on mentors' operation and
editorial videos are stored in the reference folder. The trainee can watch the reference video
at any time. So he can learn his skill without any stress of actual welding operation. Another
purpose of this system is to study relationship among voltage and amperage signals, torch
camera image, and penetrated situation. Data of more than 200 welders of several welding
companies were acquired to improve evaluation algorithm from torch camera image and
electrical signals of manual welding. The same data on automatic welding operation are also
acquired in various welding conditions. Evaluation of weld quality by one camera system
becomes possible, when the feature of the welder was stored in the data base. One camera
system is used on actual production process, and whole manual process is recorded to
evaluate the quality of the products.

Figure 34. Training machine of GTA welders of one sided full penetration butt welding.

Welding Processes 304
5. Conclusion
High speed imaging on welding process is a powerful tool to understand their nature. Deep
knowledge on principle of target process is necessary to take and analyse succesfully. Some
fundamental technique on high speed imaging are described in this paper.
Author details
Yoji Ogawa
Ogawa Giken Co.ltd. & AIST, Japan
6. References
Inoue, K. (1981) Image processing for on-line detection of weld process, Trans. JWRI, No.10,
pp.13-18
Ogawa, Y. (2008) Accurate three dimensional measurement of arc welding phenomena, J.
Soc. Instrum. Control Eng., Vol.47, pp.65-68 (in Japanese)
Ogawa, Y. (2011). High speed imaging technique Part 1, Science and Technology of Welding
and Joining, Vol.16, No.1, pp.16-26
Pattee, H. E. et al. (1973). Effect of arc radiation and heat on welders, Welding Journal,
Vol.52, pp.297-308
Shaw, C. B. Jr (1975). Diagnostic studies of the GTAW arc. Part 1. observational studies,
Welding Journal, Vol.54, pp.33s-44s
Zacharia, T. & David, S.A. (1993). Mathematical Modeling of Weld Phenomena, Welding
Journal, Vol.52, pp.3-23
Chapter 13




2012 Wglowski, licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Monitoring of Arc Welding Process
Based on Arc Light Emission
Marek Stanisaw Wglowski
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/49987
1. Introduction
A welding process may be expressed as a system shown in Figure 1 whose outputs depends
on the welding conditions or their nominal constants (which determine the dynamic model
of the system), controlled by the variables or inputs (adjustable welding parameters), and
affected by the disturbances (fluctuations or variations of welding conditions from their
nominal constants), (Zhang Y.M., 2008).

Figure 1. The welding process as a system (Zhang Y.M., 2008)
One of the major goals of the process monitoring in welding is to assure that the required
welding parameters are being applied into the process to make a quality of welds. In case
abnormal welding parameters are detected, the resultant segment of welds may be post-
examined using the more precise methods, (Blakeley P. J., 1990; Siewert T., et al, 1992; Chen
X.Q., 2002; Pan J., 2003). This would help to reduce the need of strict/expensive process
controls and reduce an extensive use of the costly post NDT (Non-destructive Testing) of all
welds. To this end, the monitoring devices are required to be fully automatic and the data
analysis of sensed signals including welding parameters and signal generated from the
welding arc need to be optimized. In addition, the monitoring devices must also incorporate
the criteria so that they can judge if the welds are acceptable or need the additional
examinations/repairs.

Welding Processes 306
Monitoring of the welding processes can be divided into the traditional and non-traditional
methods (Fig. 2). The traditional methods are based on the monitoring of the electrical and
other direct welding parameters, (Kim J.W., et al, 1991; Johnson J.A., et al, 1991; Modenesi P.
J., Nixon J.H., 1994; Luksa K., 2006). The non-traditional ones use many different signals, for
example: x-ray radiation (Guu A.C., et al, 1992), IR and UV emission (Fan H., et al, 2003),
ultrasonic wave (Carlson N.M., et al, 1992), acoustic emission (Taylor-Burge K.L., Harris T.J.,
1993) and sound (Saini D., Floyd S., 1998; Luksa K., 2003) to analyse and detect the process.
The traditional methods have been effectively used in the welding process monitoring and
control. For example, the measurements of the welding current and arc voltage can be used
to estimate the stability of the welding processes, especially with the advanced methods of
the signal analysis and AI methods, (Smith J., Lucas B., 1999). The so-called through the arc
sensing, which is based on the measurement and analysis of the welding current and arc
voltage, is a widely used traditional method which has been accepted as one of the effective
methods for the weld seam tracking. The synergic control of GMAW machines (Amin M.,
Naseer A., 1987) is also based on the measurements of the current and arc voltage. One
interesting case where on-line control of the weld quality is based on the characteristics of
the welding arc signal is the narrow groove GMAW with an electromagnetic arc oscillation
(Kang Y.H., Na S.J., 2003). The relatively complex plasma arc welding process can be also
monitored by using of an electrical signal from the pilot arc (Lu W., Zhang Y.M., 2004). In
addition, traditional methods are also useful for detecting of disturbances of the welding
process in the form of surface impurities and insufficient shield in GMAW. Monitoring is
carried out using specialized monitoring equipment or universal measurement cards.

Figure 2. Monitoring methods of welding processes and typical signals
While the traditional methods have advantages of being low-cost and have achieved many
successes as aforementioned, many existing issues may require the use of the signals more
than the welding current, arc voltage, and the other direct welding parameters. For example,
monitoring and control of a weld penetration is an important issue in welding, (Zheng B., et
al, 2009) which may require the use of the non-traditional methods. The real-time vision

Monitoring of Arc Welding Process Based on Arc Light Emission 307
systems take the lead in the non-traditional monitoring methods especially on the robotic
welding applications. The CCD (Charge Coupled Devices) video cameras which can be used
with the fast algorithms can give us real-time estimates of stability of the process, quality of
welds, for example depth of the penetration, (Zhang Y.M., et al, 1993). However, the
investment for non-traditional methods is typically high. Cost effective non-traditional
methods such as the arc light radiation monitoring which can measure and analyse the
intensity of the whole range of the arc light spectrum or intensity of a single emission line (Li
P.J., et al, 2001; Wang Q.L., et al, 1997; Yoo C.D., et al, 1997; Ancona A., et al, 2004; Sadek C.A.,
et al, 2006; Li Z.Y., et al, 2009; Mirapeix J., et al, 2008) are thus desired. This method was first
used to determine the length of the arc in the method of MAG in 1966 (Johnson C.A., et al,
1966). The spectroscopy methods are also finding application in the other welding processes
such as laser welding (Stabillano T., et al, 2009; Bruncko J., et al, 2003; Kong F., et al, 2012).
However, new methods of monitoring require the use of sophisticated measuring equipment,
which in most cases need to be adapted for the measurement in welding.
2. Arc light radiation
They are many sources of an electromagnetic radiation of the welding arc area. It can be: the
arc column, the regions close to the electrodes, the liquid metal transported across the
welding arc, the molten pool, the heated region of the base material around the molten pool,
the heated end of the electrode wire ((Pattee H.E., et al, 1973). The welding parameters
strongly influence on the range of the wavelength of the electromagnetic radiation and their
spectral composition (Hinrichs J.F., 1978). The intensity of the radiation produced by the
welding arc is a function of the welding process itself and of the welding variables. The
welding arc spectrum can be divided according to wavelength as shown in Table 1. The
welding arc in the TIG method are shown in Figure 3.

Type Wavelength [nm]
Extreme ultraviolet 4 200
Ultraviolet 200 400
Visible 400 750
Infrared 750 1300
Far infrared 1300 Hertzian wavelength
Table 1. Radiation from welding arcs (Hinrichs J.F., 1978)

Figure 3. The shape of the TIG arc (welding current 100 A, arc length 2 mm)

Welding Processes 308
The energy dissipated in the arc column is mainly dissipated by the conduction and
convection. Emission of an electromagnetic radiation is 10 to 15% of the energy supplied to
the arc (Marzec S., Janosik E., 1995). Thermal radiation, whose source is a body of high
temperature, is characterized by a continuous spectrum of radiation. The source of the
continuous spectrum in the arc is mostly a liquid weld pool (Quigley M., 1977). Radiation
characteristics of ions and atoms in the arc is a discrete (Glickstein S., 1976). This type of
radiation is analysed in the literature as a plasma radiation.
The plasma at a temperature within the range between several tens of eV and keV (the
energy scale 1 eV = 11 600 K) emits an infrared radiation, visible, ultraviolet or X-rays,
which, due to the emission mechanism can be divided into the three basic types
(Huddlestone R., 1965):
- the line radiation of atoms or ions sent during the move from the one discrete energy
level to another (transition between states related);
- the recombination radiation associated with the free-electron uptake by one of the
discrete levels of atoms or ions (the transition between the state of the free and
associated states);
- braking radiation in the free-electron zones ion (transitions between free states).
The total radiation of the plasma arc is the sum of a continuous radiation and line radiation
of the spectral lines (Szymaski A., 1991). This sum can be written as:

, , c L
c c c = +

(1)
where: ,c - intensity of radiation with a continuous spectrum, ,L - intensity of spectral
lines; sum appearing on the right side of the equation is carried out after all lines lying in the
area.
For the optically thin plasma radiation intensity with the continuous spectrum can be
written as (Szymaski A., 1991):

, ,
( ) ( ) 1 exp
c c
hc
k T B T
kT

c

| | | |
=
| |
\ . \ .
(2)
where: B(T) - the Planck function for blackbody, k,c(T) - the total absorption coefficient, T
arc temperature [K], h - Planck's constant (6,626210
-34
[Js]), c - speed of light in vacuum
(2,997910
8
[ms
-1
]), - wavelength [nm], k - Boltzmann constant 1,3810
-23
[JK
-1
].
The formula for the intensity of the continuous radiation is valid regardless of whether the
plasma is in a state of the local thermal equilibrium (LRT) or not. The intensity of a spectral
line ,L data is an expression (Szymaski A., 1991):

3
,
,
exp ( )
8 ( ) 2
q qp e i i q i
L qp
i
hcg A N N E E
h
P
U T mkT kT

c
t t
| | A | |
| = |
| |
\ . \ .
(3)

Monitoring of Arc Welding Process Based on Arc Light Emission 309
where: gq the statistical weight of upper level; Aqp probability of transition;
Ei,q ionization energy of the upper level; Ei decrease of the ionization potential; Pqp line
profile, Ne electron density, Ni joins density, Ui - the statistical sum of the ion, m mass
of particle.
A spectral distribution and an intensity of the thermal radiation depends on the body
temperature. Black bodies with the temperatures up to 500 K emit mostly the infrared
radiation at a wavelength > 2 m. Body with a temperature above about 1000 K, in addition
to a long-term infrared radiation also emit the infrared radiation close to the wavelength
range 0,78 1,4 m, and very little, because less than 1% of visible radiation. Only body at a
temperature higher than 3000 K emits infrared and a visible radiation is also a slightly
(0,1%) long-term ultraviolet radiation. Only the body with a temperature above 4000 K emits
ultraviolet radiation shorter than 315 nm (Marzec S., Janosik E., 1995). The welding arc
radiation intensity is the greatest at the wavelengths between 200 and 1300 nm [16]. The
share of the infrared radiation, visible light and ultraviolet radiation in the spectrum
depends on the welding method, and the welding parameters (H.E. Pattee, et al, 1973).
The highest intensity of the visible radiation of the arc welding processes is observed in the
MIG/MAG and next MMA, TIG and plasma welding. It was also found that the intensity of
the ultraviolet radiation increases with the square of the welding current and the intensity of
visible radiation is not growing so vigorously (H.E. Pattee, et al, 1973). The intensity of the
ultraviolet radiation and visible light emission when welding with the coated electrodes and
cored wires (MIG/MAG and self-shielding wires) in the presence of welding fumes is less
than that in case of TIG process (for similar welding current). Under the same conditions,
the intensity of the infrared radiation does not change dramatically. When a submerged arc
welding process is used the visible and ultraviolet radiation is absorbed by a layer of a flux.
Radiation characteristics of the ions and atoms in the arc has a discrete character, and the
source of radiation is mainly argon atoms and ions, iron, oxygen and nitrogen. The intensity
of radiation of the other elements is much lower. In the wavelength range of the visible
radiation spectrum iron, oxygen and nitrogen lines, and only partially spectrum of argon,
which the ionization potential is much higher are mainly composed (Petrie T.W., Pfender E.
1970). This also means, that the emission of the light by atoms and ions of argon occurs at
the higher temperatures than the temperatures reached at the arc welding for example, in a
mixture of Ar+CO2.
The discrete spectral studies provide information about the temperature of the radiation-
emitting particles, because the excitation of particles required to provide it with a certain
amount of the energy, and for this reason the temperature can be measured. The source of
this type of radiation in the arc welding is mainly plasma arc column, but also the metal
transported by the arc, slag, and the surface of the welded components (Etemadi K., Pfender
E., 1982). The energy regions close to the anode and cathode arc are consumed for heating
and melting of the electrode and the base material. It is known that the potential and kinetic
energy of the electrons are converted into the surface of the anode heat causing it to intense
heat (Petrie T.W., Pfender E., 1970).

Welding Processes 310
The arc radiation is a complex phenomenon and dependent on a number of the welding
parameters. To apply for monitoring the radiation of the arc welding process with a high
accuracy and reliability is necessary to create the model of the binding intensity of the
visible radiation from the arc welding parameters (Yoo C.D., et al, 1997). The welding arc
can be regarded as a point source of the radiation. However, this approach in many
applications seems to be insufficient. A better approach is to treat the arc as a cylindrical
source of the radiation. This model accurately reproduces the actual shape of the welding
arc and makes examination of the arc radiation easier. For this reason, it will be elaborated.
Cylindrical model can also be simplified and presented as a half-sphere of the welding arc,
which is used in the design of the monitoring systems of automated welding process, based
on the machine vision systems (Lee C.W., Na S.J., 1996; Yu J.Y., et al, 2003).
Arc column consists of the three types of particles: electrons, ions and neutral atoms. It is
assumed that the arc column is in the state of a local thermodynamic equilibrium, in which
the electron collisions play an important role in the excitation and ionization.
Equation 2 describes the arc emission of the radiation with a continuous spectrum. Given
the relationship between the wavelength and frequency c / = , and the Planck function for
the black body, as well as when h / kT << 1, the Rayleigh-Jeans'a law is performed. Then
the equation 2 can be simplified to:


2
2
2
e
k kT
c


(4)
where: - frequence [Hz], Te - the kinetic temperature of electrons [K].
Compatibility equations 2 with 4 is better than 5% for LT>4,3 cmK, where L is
wavelength [cm]. The right side of equation 4 has a value equal to 1 (approximately) for the
infrared and visible radiation. Also, at atmospheric pressure and a normal range of the
welding current, the electron temperature is close to the temperature of the arc. Considering
the above and apart from the differences in the temperature, it can be eq. 4 write as:


2
2
2
v
k v
kT
v
c

(5)
where: T the arc temperature [K].
To simplify the discussion, the gradient of the temperature along the axis of the arc can be
omitted. By combining the emission coefficients for the different areas of the arc, the energy
radiated from the entire arc can be expressed as:

dv B
iv v


(6)
After the calculation of the emission factors in the whole arc welding, and after assuming
that electrical conductivity and voltage gradient are constant and taking into consideration
the impact of visible light weld pool (Zhang Y.M., Li P.J., 2001):

Monitoring of Arc Welding Process Based on Arc Light Emission 311

2
2
3 1 4
1
2
G
I
iv
G I G LI G
e B

| |
= + +
|
\ .
(7)
where: , Gi constants, L arc length, I welding current.
Equation 7 gives a relationship between the radiation and the visible arc welding
parameters, including the current intensity and the arc length. The authors of the model
(Zhang Y.M., Li P.J., 2001) show that this equation is satisfied for arc welding with a current
of 150 A, because at the higher currents the current density is not constant over the entire
volume of the welding arc.
3. Investigation of the arc electromagnetic radiation
The investigation has been carried out to date focused mainly on examining the luminance of
the arc, impact of radiation on the health of the welders (Hinrichs J.F.,1978) and systems to
protect them, and to development the tracking systems (torch position). Analysis of the visible
light spectrum emitted by the arc welding is used to study the distribution of a temperature in
the arc (Farmer A.J.D., Haddad G.N., 1984), calculate the average temperature of the welding
arc, an amount of a hydrogen in the shielding gas (Grove L., et al, 1970), and the temperature
of molten metal weld pool. The analysis of the arc light emission may help to develop the
technique of taking photographs of the welding arc. Spectroscopic methods are a useful tool
for studying turbulent shielding gas after leaving the gas nozzle in the TIG and MIG/MAG
methods, relationship between the spectral distribution of radiation and the type of a base
material and the electron density distribution.
It should be emphasized that the study of the visible radiation in the method of the arc
welding MIG / MAG was also used to monitor the metal transfer process in the arc (Wang
Q.L., Li P.J. 1997). Methods that use the electrical signals (measure the welding current and
arc voltage) are effective only to track the short arc and globular metal transfer welding
process. When the metal is being transferred by the spray mode, the signal / noise ratio is
too small, and the greater accuracy is achieved by measuring the intensity of the visible
radiation arc (Wang Q.L., Li P.J., 1997). Optical methods are also applied to scan the length
of the welding arc in the TIG and the MIG/MAG methods.
In parallel, a wide range of plasma research is conducted. First of all, emission spectroscopy
and scattering of a laser radiation (laser spectroscopy) were used. These methods allow the
calculation of plasma parameters such as a temperature and concentration of atoms (ions,
electrons).
The emission spectroscopy is a passive method in which the electromagnetic radiation from
the plasma (one or many spectral lines) is recorded and analysed. The advantage of this
method is particularly simple measurement. This requires an optical focusing system, a
monochromator or a spectrometer and detector, which can be photomultiplier or CCD. The
disadvantage of this method is that the recorded radiation is a total emitted from the
plasma. In order to obtain measurement data from one particular point of the measurement,

Welding Processes 312
it is necessary to use the Abel transformation (Cho Y.T., Na S.J., 2005). Another
disadvantage is the need to run the calculation assumptions that the plasma is in a state of a
local thermodynamic equilibrium and is optically thin.
Laser spectroscopy is a more universal method. However, it requires a laser light source and
a detection system. The method of the laser spectroscopy allows for determination of the
plasma parameters at a given point. In some cases, calculation of the plasma parameters
without the assumption that the plasma is in thermodynamic equilibrium allowed. This
technique uses: the Rayleigh scattering, Tomson scattering, laser induced fluorescence (LIF)
and two photon laser induced fluorescence.
Plasma radiation recorded in the measurements perpendicular to the axis of discharge
(called side-on) is the sum of the smaller contributions from the various layers of plasma
(Figure 4). The known Abel transformation (Cho Y.T., Na S.J., 2005) allows to determine (x)
knowing I(x).

Figure 4. Cross section of the plasma column, the discharge axis is perpendicular to the plane of the
paper, A - radial distribution of the emission factor, B - distribution of intensity observed on the side,
I(x) - radiation distribution of intensity in the plane perpendicular to the direction in which the plasma
is observed, x distance from the direction of observation of plasma (Cho Y.T., Na S.J., 2005)
If the plasma in the observed cross-section is cylindrically symmetrical and the phenomenon
of self-absorption does not occur, the radiation distribution of intensity in the plane
perpendicular to the direction of observation of the plasma can be determined by the
formula (Cho Y.T., Na S.J., 2005):



0
2 2
2
x
r
r r
I x dr
r x

(8)
where: (r) intensity of radiation emitted by the plasma per unit thickness or distant from
the axis of the discharge, x distance from the direction of observation of plasma (Fig. 4), 2r0
diameter of the area in which the plasma occurs.

Monitoring of Arc Welding Process Based on Arc Light Emission 313
Up to now, the main goals of the plasma investigation in the welding arc were creation the
mathematical and physical models of the arc (Fan H.G., et al, 1997). These models will be
very useful to design new welding machines. Very important aspect of experiments is to
find the correlation between the electric welding parameters and the properties of the
welding arc. Many experiments concern studies about the influence the composition of the
shielding gas on the plasma properties. Also some investigation concern a magnetic arc
deflection (Kang Y.H., Na S.J., 2002). The most important aim of investigation is to calculate
the arc efficiencies. Many experiments were focus on the physical properties of the plasma
welding arc, for example the temperature distribution, velocity fields of the electrons and
ions, electrode work functions, and the local thermodynamic equilibrium in free-burning
arcs in argon. Modern methods of welding, including A-TIG method, prompted the author
(Ogawa, Y. 2004) to study the effect of the additional elements and compounds intentionally
introduced to the area of the welding arc on properties.
In a study of the arc radiation it is important to determine the influence of individual factors
on the width of the spectral peaks. Typical spectral line profile is shown in Figure 5 together
with the characteristic values: xc wavelength of the center line, FWHM - Full Width at Half
Maximum, Imax maximum value for the radiation intensity of spectral line. The natural
width of the spectral lines (Huddlestone R.H., Leonard S.L., 1965) is due to the finite lifetime
of the energy levels and is higher, the lifetimes are shorter. Emission line profile resulting
from natural broadening is the Lorentz distribution.

Figure 5. Typical spectral line profile
The second important factor is the Doppler broadening of the spectral lines, which is
associated with the movement of the particles emitting the radiation. If the emitter has a
velocity component of the direction consistent with the observation, the relative change in
wavelength, involve a change in frequency is called Doppler effect. In the case of a thermal
motion when emitting particle velocity distribution is Maxwell's distribution, the profile of
the emitted spectral line is the Gaussian profile (Zieliska S., 2004).
Another kind of broadening, which can be encountered in the analysis of spectral lines, is a
pressure broadening. This kind of broadening of the spectral line is the result of collisions

Welding Processes 314
with the other particles emitter. They can limit the lifetime of the excited atomic levels, and
thus lead to an broadening of the line profile, which in this case is the Lorentz distribution.
There are basically three types of the pressure broadening: the resonances, van der Waals
and Stark (Zieliska S., 2004).
Components of the measurement system is a factor caused the further broadening of the
spectral line. Apparatus profile in this case is the Gaussian profile. Theoretically, the
spectrometer apparatus function should be linearly dependent on the wavelength. In
reality, however, the profile apparatus is a convolution of functions associated with the
matrix detector and functions of the optical elements of the spectrometer (Zieliska S.,
2004).
The factors leading to the broadening of the spectral lines can be divided into given the
Lorentz and Gaussian profiles. Their impact on the value of the broadening is different and
may depend on conditions in the plasma. Resultant spectral line profile is a function which
is a convolution of the Lorentz and Gauss functions (Huddlestone R.H., Leonard S.L., 1965)
called the Voigt profile.
It should be noted that the photo-detector (CCD detector), in addition to the signal,
measures also the background radiation. To eliminate the influence of background
radiation, when analysing the intensity distribution of the welding arc radiation this
radiation must be subtracted.
Taking into account the resolution of the transmitters, the recorded peaks are "cluster" of
several spectral lines of a single element (Fig. 6), or even a few elements in the various
degrees of ionization. So, matching the shape function is important in determining the
exclusion of gravity for that group of peaks. The first element of the monitoring system of
welding processes is to develop methods for the identification and measurement of the
characteristic quantities of recorded spectral line, such as a peak width, position and
amplitude of maximum. The examination which function better describes the profile of the
peak ("cluster" of spectral lines) seems to be crucial for the detection of disturbances of the
welding process. Profiles of the peaks can be matched using Gaussian, Lorentz and Voigt
functions (Fig. 6).
Matching functions are carried out mostly using the least squares method (eg. Levenberg-
Marquardt algorithm) and special software can be adapted for this purpose. On the basis of
the matching function parameters the position of maximum spectral line (xc) and the
spectral line width (FWHM) can be determined. In developing the experimental data even
in so-called matching additive constant y00 must be considered. The constant is present due
to the additional signals recorded by the measurement apparatus.
From the equation (7), a relationship between intensity of the welding arc radiation
(energy radiated for a given spectral line xc) Biv, arc length and welding current can be
designated. Using a software, coefficients Gi and can be estimated basing on the
collected data.

Monitoring of Arc Welding Process Based on Arc Light Emission 315

Figure 6. An example of peak matched Gaussian functions, Lorentz and Voigt, I=200 A, L=3 mm, 100 %
Ar with marks of spectral lines .
The values of the parameters characterizing the welding process can be determined
minimizing the sum of squares:

( )
( ) { } ( )
2
2
2
1
, , , , ,
kl k l kl k l i
l k
kl
B I L B I L g
B
_
(
=

A


(9)
where: Ik welding current; Ll arc length; Bkl(Ik, Ll, ) - intensity of light of wavelength
recorded during the welding current Ik, at arc length Ll; Bkl - uncertainty set of light
intensity Bkl(Ik, Ll, );
kl
B

(Ik, Ll, , {gi}) - defined by formula (7) the theoretical intensity of the
light with a wavelength recorded during the welding current Ik, at arc length Ll; {gi}={G1, ,
G2, G3, G4} - set of values of the parameters appearing in formula (7).
Uncertainty-determination and the parameter g e {g} = {G1, , G2, G3, G4} is determined by
the method described in (Koczak S., Nowak M., 1981):

2
1
i ii
p
h
m m
_
c

=

(10)
where:
1
ii
h

- ii component of the inverse Hesse matrix; _


2
- sum of squared deviations from
the theoretical value of the experimental results; mp- number of experimental results; m-
number of parameters designated by the matching.
The components of the Hesse matrix model is defined as (Koczak S., Nowak M., 1981):

( )
( )
{ } ( ) { } ( )
{ } ( )
2 2
2
, ,
2
,
, , , , , ,
, ,
1
2
, , ,
ij
i j
kl teor k l i kl teor k l i
kl k l
i j i
l k
kl teor k l i
kl
j
h
g g
B I L g B I L g
B I L
g g g
B I L g
B
g
_

c
= =
c c
(
c c
(

c c c (
= (
c (
A
(
c
(

(11)

Welding Processes 316
4. Application of the arc light emission to monitor the welding process
In order to monitor the welding processes successfully, the optical sensing systems have
been developed. Special procedures, models and modification of the monitoring devices
have to be implemented with the sensing systems, some of the systems are discussed herein.
Some typical application examples included sensing of the arc length in the TIG welding
(Wglowski M.St., 2010), relationship between the welding conditions and intensity of the
arc light emission in the MIG/MAG methods (Wglowski M.St., 2008; Wglowski M.St.,
Zhang Y. M., 2010) and the influence of the parameters and disturbance of the welding
process on the shape of the spectrum of the arc light radiation (Wglowski M.St., 2009). In
this part, their principles are being described.
4.1. Sensing of the arc length in the TIG welding method based on the arc light
intensity
One of the main task of the monitoring systems in the robotized and automated welding
stations is the measurement and control of the arc length. The main objective of the
investigation was to study the possibilities of using of the visible radiation of the welding
arc for stability monitoring of the TIG welding process, giving consideration to the changes
of the intensity of visible light radiation with the changes of the welding current or welding
arc length (reproducing the case of burn- through and arc migration). The arc length is one
of the basic welding parameter in the TIG method, which directly influences the arc voltage.
The arc length has an effect on the distribution of an arc energy, and as the consequence on
the amount of heat put into the welded joint and on the width of the weld.
The tests have been performed on the stand for the automatic TIG welding. The measuring
system consists of welding current and voltage transducers, an electrooptical converter, a
measurement card and a PC computer. The analysed beam of the visible radiation is fed into
the electrooptical converter by means of a standard optical wave guide. The electrical signal
corresponding to the visible light intensity and signals from the welding circuit are recorded
on the PC by the recording device, equipped with the NI DAQ 6036 measuring card. The
recorded signals were then analysed. The intensity of the visible light radiation of the
welding arc was measured in volts. The following experiment conditions were approved:
the arc burns between the thoriated tungsten electrode (cathode) and a copper plate (anode),
the welding torch is fixed, argon (Ar) as the shielding gas (gas flow rate qg=10 dcm
3
/min),
welding current source: Kemppi Pro 5000 (DC current set in the range of 30300 A). It was
assumed that the arc length is equal to the distance between the electrode tip and the
welded metal surface. The range of the welding arc length L=25 mm. In Figure 7 show the
configuration of the optical system relative to the welding torch is shown.
The arc length was changed in the range from 2 to 5 mm during the experiments. Figure 8
shows the influence of the arc length L on the visible light intensity and arc voltage for the
welding currents in the range of 50300 A. It can be seen that considerable changes of the
arc length are followed by the substantial changes of radiation intensity of the welding arc
(wave length 696 nm) and only by small changes of the arc voltage.

Monitoring of Arc Welding Process Based on Arc Light Emission 317

Figure 7. Configuration of the optical system relative to the welding torch
Three cases of transient states of the welding arc length have been also investigated: the
abrupt change of the arc length (Fig. 9a), the abrupt change of the welding arc length,
simulating the burn-through of the joint (Fig. 10a) and a smooth change of the welding arc
length, simulating a bad preparation of welded elements or their distortion during welding
(Fig. 11a). These are typical transient states in the welding practice.
The abrupt change of the welding arc was forced by a proper preparation of the 20 mm thick
plate by milling (Fig. 9a). The height of the received steps was 1 and 2 mm, which resulted
in the arc length of 1, 2 and 4 mm at a welding current of 100 A (DC). Results of the
measurement performed at the welding speed of 60 cm/min are presented in Figure 9b. The
moment of entering the step by the welding torch is shown by arrows.

Figure 8. Influence of welding arc length on the light emission (Biv) and the arc voltage (Np) at the
welding current in the range of 50300 A. Argon as shielding gas
The second tested transient state was the abrupt change of the arc length simulating the burn-
through of the welded joint. A 20 mm thick plate were prepared by drilling holes with a diameter
of 1,36 mm (Fig. 10a). The arc length during the experiment was maintained at 3 mm at the
3

Welding Processes 318

Figure 9. (a) scheme of the experiment with the forced abrupt change of welding arc length, (b)
measurement results of the welding current, arc voltage and intensity of arc radiation at abrupt changes
of the welding arc length, welding speed 60 cm/min
welding current of 100 A (DC) and a welding speed of 60 cm/min. The test results are
presented in Figure 10b. The moment of entering the holes by the welding torch is shown by
arrows.
The third tested transient state was a smooth change of the arc length simulating the
deformation of welded plates or improper preparation of the joint. Plates 20 mm in
thickness were welded at the angle of 5 (Fig. 11a). The arc length during the experiment
changed in the range of 17 mm at the welding current of 100 A (DC) and a welding speed
of 60 cm/min. The test results are presented in Figure 11b.

Figure 10. (a) scheme of the experiment with the forced abrupt change of welding arc length simulating
the burn-through of the welded joint, (b) measurement results of the welding current, arc voltage and
intensity of arc radiation at abrupt changes of the welding arc length simulating the burn-through of the
welded joint, welding speed 100 cm/min,
The one of the most important factor is the influence of the changes of welding current and
the arc length during TIG welding on the intensity of the visible radiation. On the basis of
the collected test data and equation 2 a relationship combining the intensity of the welding
arc radiation (Biv) with the arc length (L) and the welding current intensity (I) can be
determined. The arc length is in the range of 2 5 mm. Based on the formula (7) presented
(a)
(b)
(a) (b)

Monitoring of Arc Welding Process Based on Arc Light Emission 319
in the paragraph the relationship can be determined. Using a software, basis of the collected
data and formulas (9-11), coefficients Gi and can be estimated.

Figure 11. (a) scheme of the experiment with the smooth change of the welding arc length simulating
improper preparation of the joint or deformation of welded plates, (b) measurement results of the
welding current, arc voltage and intensity of the arc radiation at the smooth change of the welding arc
length, welding speed 60 cm/min
Two cases were taken into account during calculation:
- theoretical Zhang model, in this model coefficient =2 and then it can be written (eq. 7):

2
2 2
3 1 4
1
2
G
I
iv
G I G LI G
e B
| |
= + +
|
\ .
(12)
- generalization model - coefficient is a parameter dependent on the measured data.
Basing on the formula (9) the calculation were carried out taking into account the following
two cases:
- Bkl - uncertainty set of the light intensity is constant for all data and it is not taken into
account during calculations; in this case the fitting will be worse for smaller values,
- Bkl - uncertainty set of the light intensity is not constant for all data and it is taken into
account during calculations.

No Coefficients Theoretical model Generalization model
1 G1 3,8(2)10
-5
1,11(1)10
-3

2 G2 56(3) 9(2)
3 G3 -4,4(1)10
-5
-3,98(12)10
-5

4 G4 1(57)10
-3
-1,8(6)10
-1

5 2 1,455(4)
6 _
2
sum of the least-squares of the deviations 7,33 3,6
7 correlation coefficient R2 0,98 0,99
Table 2. Results of calculation of coefficients Gi for theoretical and generalization models at arc length
in the range of 2 - 5 mm, Bkl constant
(a)
(b)

Welding Processes 320

No Coefficients Theoretical model Generalization model
1 G1 4,6(2)10
-5
1,7(1)10
-3

2 G2 34(2) 2(65)10
-2

3 G3 -4,06(10)10
-5
-35,1(6)10
-6

4 G4 -1,09(16)10
-1
-11,4(9)10
-2

5 2 1,364(2)
6 _
2
sum of the least-squares of the deviations 5,75 1,07
7 correlation coefficient R
2
0,99 0,99
Table 3. Results of calculation of coefficients Gi for theoretical and generalization models at arc length
in the range of 2 - 5 mm, Bkl is not constant
Taking into account the results given in Tables 2 and 3, the sum of the least-squares of the
deviations is smaller for the generalization model and for case were weight Bkl is not
constant. Finally the formula 9 can be written as:

0,02
2
1
1, 364
0,000035
0, 0017 0,114
2
I
I
LI e B
iv
| |
|
=
|
|
\ .
(13)
This equation is satisfied for the wavelength 698 nm and the arc length in the range of 2 - 5
mm. The graphic presentation of this formula is shown in Figures 12a and 12b. The arc
burns between the thoriated tungsten electrode (cathode) and a water cooled copper plate.

Figure 12. (a) relationship between intensity of the welding arc radiation in the TIG method and
welding current at arc length 2 mm, and the wavelength 698 nm, (b) relationship between an intensity
of the welding arc radiation in the TIG method and the arc length, welding current at arc length in the
range of 2-5 mm, and the wavelength 698 nm
The change of the TIG welding arc length causes also changes of the intensity of the visible
arc radiation. An increase of the arc length results in the intensity increase of the selected
spectral line (696 nm) of the TIG welding arc visual radiation. This increase depends on the
welding current intensity. Larger increases of the visual arc radiation intensities are
observed at the higher welding currents.

Monitoring of Arc Welding Process Based on Arc Light Emission 321
Three cases of the welding arc length transient states have been tested. It has been found that
in all that cases the intensity of the arc radiation at the wave length of 696 nm is much more
sensitive to abrupt changes of the welding arc length, than the arc voltage. In the first case (Fig.
9) the change of the arc length for 1 and 2 mm was followed by a 400% change of the radiation
intensity and only by a 10% change of the arc voltage. In the second tested case, simulating the
burn-through of the welded joint, considerable changes take place for both - the arc voltage
and intensity of the arc radiation, but in the radiation intensity record the peaks corresponding
to the consecutive holes (Fig. 10) can be more easily identified. By the arc voltage measurement
a hole with a diameter of 4 mm can be identified, while the measurement of radiation intensity
makes possible the identification of a 1,3 mm hole. The third tested transient state was a
smooth change of the welding arc length, simulating the deformation of welded plates or
incorrect setup for welding. Also in that case the changes of the welding arc length are
followed by considerable changes of radiation intensity and smaller changes of the arc voltage.
4.2. Relationship between the welding conditions and intensity of the arc light
emission in GMAW
This section describes the acquisition and analysis of the arc light emission and its correlation
with the welding parameters and disturbances of the welding process. A spectrophotometer
card PCI 2000 ISA-A in the visible spectral range of 340 nm to 860 nm was used in the study.
The measurement system consisted of the welding current and voltage transducers, an electro-
optical converter, a data acquisition card and a PC computer (Fig. 13). Signals from the
welding circuit were recorded on the PC through the data acquisition card NI DAQ 6036. The
measurements during bead-on-plate welding and joints welding were carried out. The signals
were analyzed in time domain. During trials, a spectrophotometric card PCI 2000 ISA-A,
which has been designed for the CCD Sony model ILX511 detector in the visible spectral range
of 340 nm to 860 nm was used to image and record the arc light spectrum for the later analysis.
The CCD detector was a line scan array of 2048 pixel.

Figure 13. Experimental setup with the flow of data

Welding Processes 322
The work-piece was moved while torch was in a fixed position such that the arc light sensor
was stationary in relation to the work-piece. The sepctrophotometric card used a sampling
time 3 ms. An optical system was used to focuse the welding arc light. The entire arc column
has been analyzed as a single object. During welding the arc voltage, the welding current,
the wire feed speed and the intensity of the arc light emission were continuously measured.
The Hall effect current sensor Model PR 1001 was used to measure the welding current. This
sensor provides electrical isolation between the current carrying conductor and the output
of the sensor. The voltage was measured by a resistance bridge by a LV 25 P transducer in
the output of the power supply. The wire feed speed was measured by E21 MPL10
transducer. The intensity of the arc light emission was measured by PIN BPW34
photodiode. Signals from the welding circuit were recorded on the PC through the IPP-2
measured system designed in the Instytut Spawalnictwa (Institute of Welding), witch based
on SCXI data acquisition system National Instruments. This system consisting of the
National Instruments SCXI-1125 is 8-channel isolated analog input modules and data
acquisition board NI DAQ 6036 E. Whole system was placed in the SCXI-1000 chassis. The
signals were recorded at a sampling rate of 20 kHz.The torch was moved at the travel speed
25 cm/min to make bead-on-plate welds and weld. Direct current levels between 104 A and
235 A were examined, all at an operating voltage in the range of 16.5 V - 25.5V. Figure 14a
shows the arc spectrum obtained in the range of 360-860 nm at the welding current in the
range of 104 235 A. The graph is presented in a logarithmic scale.
As shown in Figure 14a the increase of the welding current causes increase of the arc light
intensity in the whole range. The shape of the spectrum was modeled by the three
mathematic functions: Lorentz, Gausse and Voight (Fig. 14b). The fitting for both single
wavelength and multiple wavelengths was carried out mathematically and the best result
was achieved with the Lorentz function. The central wavelength, intensity and FWHM - Full
With at Half Maximum were calculated. The main source of the arc light radiation in the
GMAW is liquid metal. The lines from the shielding gases have not been found. The
detailed analysis of influence of the welding current on the arc light spectrum was
previously discussed (Wglowski M.St., 2008, 2009).

Figure 14. (a) effect of the welding current on the arc light spectrum. The welding current in the range
of 104-235 A, Ar + CO2 as the shielding gas, wavelength in the range of 480-860 nm. Logarithmic scale,
(b) calculated line profile of the wavelength 439,28 nm compared with measured values and the
Gaussian, Lorentz and Voight functions

Monitoring of Arc Welding Process Based on Arc Light Emission 323
The purpose of these studies was also to check the influence of disturbances of the welding
process on the arc light intensity. To this end, arc light intensity was measured during welding
of real joint, 4 mm in thickness. The disturbances of the welding process was tha additional
filler metal in the grove. The experiments were done under the following conditions: welding
current I=160 A, arc voltage U=21.2 V, shielding gas M21 Ferromix C18, welding speed 25
cm/min), wire EN 440 G3Si1, base material S235, the grove was prepared for Y.
The weld produced is shown in Figure 15a. The scheme of the method of disturbance of the
welding process is shown in Figure 15b. The macroscopic examination of the padding welds
are shown in Figure 16. During welding the arc voltage, welding current, wire feed speed
and the intensity of the arc light emission were continuously measured. The intensity of the
arc light signal is shown in Figure 17.

Figure 15. (a) welding joint with marked areas of disturbances of the welding process, (b) scheme of
disturbance of the welding process

without disturbances disturbance 1 disturbance 2
Figure 16. Macroscopic examination of the padding welds. Etching Adler, magnification x2

Figure 17. Intensity of the arc light signal recorded during welding of plate with disturbance
(a)
(b)
(a) (b) (c)

Welding Processes 324
To estimate the stability of the welding process based on the arc light emission the least
squares method was used. To model the arc light signal a cubic polynomial was used:

2 3
0 1 2 3
y a a x a x a x c = + + + + (14)
where: y arc light emission, x time, ai coefficients, residual the differences between
the observations and the model.
To calculate a residual the following formula should be solved:

(
.
( )
.
(
y m M
Y y m
y m M
(
(
(
(
=
(
(
(
+


2 3
2 3
2 3
1 ( ( ) ( )
. . . .
1 ( ) ( ) (
. . . .
1 ( ) ( ) (
x m M x m M x m M
X
x m x m x m M
x m M x m M x m M
(

(
(
(
=

(
(
(
+ + + (


0
1
2
3
a
a
a
a
|
(
(
(
=
(
(
(

(15)
where: Y- intensity of the arc light signal matrix, X time matrix, coefficients matrix, m
discrete time.
But this system is over determined. There are more equations than unknowns. So it cannot
expect to solve the system exactly. Instead, it can be solved it in the least squares sense:

min X Y
|
|
(16)
A theoretical approach to solve the over determined system begins by multiplying both sides
by XT. This reduces the system to a square, n-by-n system known as the normal equations:

T T
X X X Y | =
(17)
If there are thousands of observations and only a few parameters, the design matrix X is
quite large, but the matrix X
T
X is small. It has been projected Y into the space spanned by
the columns of X. Continuing with this the theoretical approach, if the basis functions are
independent, then X
T
X is nonsingular and

( )
1
T T
X X X Y |

= and Y X| = (18)
then, the residual can be calculated as:

( ) ( )
2
2 1
T
Y Y Y Y
M
c o

= =
+
(19)
where: 2M+1 the number of data used in the fitting.
To calculate formulas 15 and 19 the following parameters were established: m in the range
of 10000 to 590000 at the step 100 and M=10. To estimate the points of unstability of the

Monitoring of Arc Welding Process Based on Arc Light Emission 325
welding process, model by polynomial can be used. To estimate the best degree of the
polynomial the F-test method can be used. Based on the previously least squares
methodology, the residual of polynomials P can be calculated as:


2
T
P
Y Y
n



(20)
where: residual from eq. 7, Y value of polynomial function, n the number of data.
The calculation acc. to formula 20 were carried out for 50 different polynomials. The best
results can be achieved for polynomial 28
th
degree (Figure 18).

Figure 18. The results of model by polynomial 28
th
degree with marked the areas with disturbances.
It is shown that the arc light signal can be utilized to monitor the welding processes. This
signal is sensitive to any changes in the welding area. The spectrophotometric card can be
useful tool to investigate the properties of the welding arc.
4.3. The influence of the parameters and disturbance of the welding process on
the shape of the spectrum of the arc light radiation
The tests were performed on the stand for automatic MAG welding operations by the
control consol. The testing plate for welding was fixed while the welding was moved at a
controlled speed. The torch was located perpendicularly to the welding surface. All
experiments were performed by the bead-on-plate welding. The measuring system
consisted of the welding current and the voltage transducers, a PC computer equipped with
the CCD spectrophotometer card, and a speed wire measurement device. The electrical
signals from the current and voltage transducers were recorded on the PC equipped with
the NI DAQ 6036 measuring card. The analyzed radiation was fed into the CCD
spectrophotometer by means of a standard fibre optics. A spectrophotometer card PCI 2000
ISA-A (Ocean Optics Inc.) was used in this research. The output of each pixel is converted to
an electrical current which represents the amount of the energy that has fallen on each pixel
in a relative manner.

Welding Processes 326
Figure 19 shows the different influence of the welding current intensity on the amplitudes,
additive constants and widths (FWHM) of the Lorentz functions that fit the best spectral
peaks of the arc light. One can see that the influence manifests in different ways in the cases
of the different spectral peaks. Generally, the welding current intensity strong influences the
additive constants and amplitudes of the peaks in the spectral range from 400 nm to 500 nm.
Figure 19d shows the dependence of the best fitted additive constants on intensity of the
current in the welding process of the clean mild steel S 235 (I=104 A; U=16,5 V).

Figure 19. (a) influence of the welding current intensity on the amplitudes, (b) the values of additive
constant in formula (3), (c) FWHM and (d) intensity of single emission lines,
Investigations on the effect of imposed disturbances, in the form of paint or grease layers
on the plate surface, on the intensity of the MAG arc light radiation in the visible range,
have been performed. The mild steel S 235 as the welding plate and the 1,2 mm diameter
SG2 type welding wire were used. Experiments were performed with shielding gas 82% Ar
and 18 % CO2. The plate surface was clean, covered with oil paint or covered with a
machine grease. It was found that both the paint and grease layer influence the recorded
spectral characteristics of the MAG welding light radiation. Figure 20 presents the different
influence of the existence of paint on the welded plate on the amplitudes, additive
constants and the widths of the Lorentz functions that fit the best spectral peaks of the arc
light. One can see that the influence manifests in different ways in the cases of the different
spectral peaks.
(a)
(b)
(c)
(d)

Monitoring of Arc Welding Process Based on Arc Light Emission 327
The presented investigations show that the low resolution spectral characteristics of the arc
light emission registered with CCD device can be applied for the purpose of monitoring of
the welding process. The arranged measuring stand has made it possible to record the
visible spectrum of the radiation of the welding arc within the range of wavelengths from
380 nm to 780 nm. The measuring stand comprised a spectrophotometer, a computer
recording the results of the measurements and a device for mechanized welding.
It was found that the spectral distribution of a single peak in the low resolution spectral
characteristics can be best fitted with the Lorentz function. In the recorded spectrum of the
welding arc light emission, separation of the ionic or atomic lines is not possible. However,
the correlation between the parameters of the fitted Lorentz function and welding
parameters (i.e. welding current) was obtained. The Lorentz function parameters depend
also on the disturbances in the MAG welding process, e.g. their values are different in the
cases of clean and painted surface of the welded mild steel S 235 plate.

Figure 20. (a) the influence of disturbances on the MAG welding arc light spectrum (I=169 A, U=19,7 V)
(b) influence of paint on the welded plate on the amplitudes, (c) the values of additive constant in
formula (3) and (d) FWHM (Attention: curves presented in the figure cannot be interpolated)
5. Summary
Modern monitoring methods of the welding processes are inherent in each automatics and
robotics production system. These systems detect very rapidly any incorrectly made weld
(a)
(b)
(c)
(d)

Welding Processes 328
joints during manufacturing and this way decreases the costs of production. That means
possibility of detecting of any faulty parts without very costly nondestructive examinations.
At present very popular conventional monitoring methods of the welding processes, based
on measurements of the welding current and arc voltage in many cases are inefficient and
are replaced or/and completed by nonconventional monitoring methods.
One of the most popular nonconventional monitoring method is sensing system based on
the arc light emission. The main aim of these investigations was to check the possibilities of
applying the visible radiation of the welding arc for the purpose of monitoring of the quality
of the welding process.
The arranged measuring stand made it possible to record the visible spectrum of the
radiation of the welding arc within the range of the wavelengths from 380 nm to 780 nm.
The measuring stand comprised a spectrophotometer, a computer recording the results of
measurements and a device for mechanized welding.
Results of the performed investigations in the field of measurement of light radiation
intensity during TIG and MIG/MAG welding, shown in this chapter, indicate that this signal
can be used for the monitoring of the welding process quality. The experience gained during
these investigations allows for further research on the welding arc radiation phenomenon.
The obtained knowledge increases the possibilities of using the signal for on-line monitoring
of the welding process on the automated and robotized stands. The analysis of the spectrum
of the welding arc radiation should help to develop the new vision sensor in the arc
welding.
The investigations are continued in the many research centers, and cover the following
issues:
- utilize the artificial intelligence method to estimate the stability of the welding process,
- develop the filtering method, and methodology to signal analysis in time, and
frequency domain,
- laser diagnostic on the welding arc in the TIG and MIG/MAG methods, develop
method of measurement of the arc light emission in many points simultaneously, and
many others.
Author details
Marek Stanisaw Wglowski
Instytut Spawalnictwa (Institute of Welding), Poland
6. References
Zhang, Y.M. (2008). Real-time weld process monitoring. Woodhead Publishing in Materials,
ISBN 978-1-84569-268-1, Cambridge, England
Blakeley, P. J. (1990). Who should be using arc monitoring? Welding and Metal Fabrication,
Vol. 58, No 6, pp. 268-272, ISSN 0043-2245

Monitoring of Arc Welding Process Based on Arc Light Emission 329
Siewert, T., Madigan, B. (1992). Through the arc sensing for measurement Gas Metal Arc
weld quality in real time. Materials Evaluation, Vol. 50, No 11, pp. 1314-1318, ISSN 0025-
5327
Chen, X. Q. (2002). Advanced automation techniques in adaptive material processing, Word
Scientific Publishing, ISBN 981-02-4902-0, Rever Edge, New Jersey, USA.
Pan, J. (2003). Arc welding control. Woodhead Publisher Limited, ISBN 0-8493-1772-X,
Cambridge, England.
Kim, J. W., Na, S. J. (1991). A study on an arc sensor for Gas Metal Arc welding of horizontal
fillets. Welding Journal, Vol. 70, No 8, pp. 216s-221s, ISSN 0043-2296
Johnson, J. A., Carlson, N. M., Smartt H.B., et al (1991). Process control of GMAW: sensing of
metal transfer mode. Welding Journal, Vol. 70, No 4, pp. 91s-99s, ISSN 0043-2296
Modenesi, P. J., Nixon, J. H. (1994). Arc instability phenomena in GMA welding. Welding
Journal, Vol. 73, No 9, pp. 219s-224s, ISSN 0043-2296
Luksa, K. (2006). Influence of weld imperfection on short circuit GMA welding arc stability.
Journal of Materials Processing Technology, Vol. 175, pp. 285-290, ISSN 0924-0136
Guu, A. C., Rokhlin, S. I. (1992). Arc welding process control using radiographic sensing.
Materials Evaluation, Vol. 50, No 11, pp. 1344-1348, ISSN 0025-5327
Fan, H. Ravala, N. K., Wikle III, H. C., et al. (2003). Low-cost infrared sensing system for
monitoring the welding process in the presence of plate inclination angle. Journal of
Materials Processing Technology, Vol. 140, No 1-3, pp. 668-675, ISSN 0924-0136
Carlson, N. M., Johnson, J. A. (1992). Ultrasonic NDT methods for weld sensing. Materials
Evaluation, Vol. 50, No 11, pp. 1338-1343, ISSN 0025-5327
Taylor-Burge, K. L., Harris, T. J. (1993). The real time analysis of acoustic weld emission
using neural networks, Proc. of the International Conf. on the Joining of Materials JOM-6
ed. Rasmus Knudsens Vej, pp. 60-67. Helsingor, Denmark
Saini, D., Floyd, S. (1998). An investigation of Gas Metal Arc welding sound signature for
on-line quality control. Welding Journal, Vol. 77, No 4, pp. 172s-179s, ISSN 0043-2296
Luksa, K. (2003). Correspondence between sound emission generated in GMA [MIG/MAG]
welding process and signals registered in the arc circuit. Welding International, Vol. 17,
No 6, pp. 438-441, ISSN 0950-7116
Smith, J., Lucas, B. (1999). Putting intelligence into welding rule based systems, fuzzy logic
and neural networks. Welding & Metal Fabrication, Vol. 67, No 10, pp. 7-9, ISSN 0043-
2245
Amin, M., Naseer, A. (1987). Synergic control in MIG welding parametric relationship for
steady DC open arc and short circuiting arc operation. Metal Construction, Vol. 19, No 1,
pp. 22-28, ISSN 0026-0541
Kang, Y. H., Na, S. J. (2003). Characteristics of welding and arc signal n narrow groove
GMAW using electromagnetic arc oscillation. Welding Journal, Vol. 82 No 5, pp. 93s-99s,
ISSN 0043-2296
Lu, W., Zhang, Y. M. (2004). Sensing of weld pool surface using non-transferred plasma
charge sensor. Measurement Science and Technology, Vol. 15, No 5, pp. 991-999, ISSN
0957-0233

Welding Processes 330
Zheng, B., Wang, H. J. (2000). Control for weld penetration in VPPAW of aluminum alloys
using the front weld pool image signal. Welding Journal. 70 (12): 363-s to 371-s, ISSN
0043-2296
Zhang, Y. M., Wu, L. (1993). Determination joint penetration in GTAW with vision sensing
of weld face geometry. Welding Journal, Vol. 72, No 10, pp. 463s-469s, ISSN 0043-2296
Li, P. J., Zhang, Y. M. (2001). Precision Sensing of Arc Length in GTAW Based on Arc Light
Spectrum. Journal of Manufacturing Science and Engineering, Vol. 123, No 1, pp. 62-65,
ISSN 1087-1357
Wang, Q. L., Li, P. J. (1997). Arc light sensing of droplet transfer and ist analysis in pulsed
GMAW processes. Welding Journal, Vol. 76, No 11, pp. 458s-469s, ISSN 0043-2296
Weglowski, M. St. (2007). Utilization of the arc light emission emitted during TIG welding to
monitoring this process. Archives of Mechanical Technology and Automation PAN, Vol. 27,
No 1, pp. 103-112. (in Polish), ISSN 1233-9709
Yoo, C. D., Yoo, Y. S., and Sunwoo H.K. (1997). Investigation on arc light intensity in gas
metal arc welding. Part 1: relationship between arc light intensity and arc length.
Proceeding of the Institution of Mechanical Engineers. Part B: Journal of Engineering
Manufacture, Vol. 211 (B5), pp. 345-353, ISSN 0954-4054
Yoo, C. D., Yoo, Y. S., and Sunwoo H. K. (1997). Investigation on arc light intensity in gas
metal arc welding. Part 2: application to weld seam tracking. Proceeding of the
Institution of Mechanical Engineers. Part B: Journal of Engineering Manufacture, Vol. 211
(B5), pp. 355-363, ISSN 0954-4054
Ancona, A., Lugara,P.M., Ottonelli,F., Catalano, I.M. (2004). A Sensing Torch for On-line
monitoring of the Gas Tungsten Arc Welding Process of Steel Pipes. Measurement
Science and Technology, Vol. 15, No 12, pp. 2412-2518, ISSN 1361-6501
Sadek, C.A., Diogo, D.S.M., Marcelo, S.M. (2006). Emission Spectrometry Evaluation in Arc
Welding Monitoring System. Journal of Materials Technology, Vol. 179, No 1-2, pp. 219-
224, ISSN 0924-0136
Li, Z. Y., Wang, B., Ding, J.B. (2009). Detection of GTA Welding Quality and Disturbance
Factors with Spectral Signal of Arc Light. Journal of Materials Processing Technology, 2009,
Vol. 209, No 10, pp. 4867-4873, ISSN 0924-0136
Mirapeix, J., Cobo, A., Fernandez, S., Cardoso, R., Lopez Higuera, J.M. (2008). Spectroscopic
Analysis of the Plasma Continuum radiation for On-line Arc-welding Defect Detection.
Journal of Physics D: Applied Physics, Vol. 41, No 3, pp. 135202135210, ISSN 1361-6463
Johnson, C.A., Sciaky, A.M. (1966). System for controlling length of welding arc, Patent US, nr
3236997
Zhang, Y.M., Li, P.J. (2001). Precision sensing of arc length in GTAW based on arc light
spectrum. Journal of Manufacturing Science and Engineering, Vol. 123, No 2, pp. 62-65,
ISSN 1087-1357
Garcia-Allenda, P.B., Mirapeix, J., Conde, O.M., Cobo, A., Lopez-Higuera, J.M. (2008). Arc
welding spectroscopic monitoring based on feature selection and neutral networks.
Sensors, Vol. 8, No 10, 6496-6506, ISSN 1424-8220
Pattee, H.E, Myers L.B., Evans R.M., Monroe R.E. (1973). Effect of Arc Radiation and Heat
on Welders. Welding Journal, Vol. 52, pp. 297s-308s. ISSN: 0043-2296

Monitoring of Arc Welding Process Based on Arc Light Emission 331
Hinrichs J.F., (1978). Radiation and arc welding: New data to enhance safety. Welding and
Metal Fabrication, Vol. 51, pp. 102-103, ISSN 0043-224
Marzec, S., Janosik, E. (1995). Influence of arc radiation of health of welder Wpyw
promieniowania uku spawalniczego na organizm spawacza. Builetyn of Welding
Institute, Vol. 39, No 6, 33-37, ISSN 0867-583X
Quigley, M. (1977). Physics of the welding arc. Welding and Metal Fabrication, Vol. 9, No 12,
pp. 619-625, ISSN 0043-2245
Glickstein, S. (1976). Temperature measurements in a free burning arc. Welding Journal, Vol.
55, No 8, pp. 222s-229s, ISSN 0043-2296
Huddlestone, R.H., Leonard, S.L. (1965). Plasma diagnostic techniques. Academic Press, New
York, USA
Szymaski, A. (1991). Spectroscopic and Computer Modeling Investigations of a Laser-Sustained
Plasma. Institute of Fundamental Technological Research Works, Polish Academy of
Sciences, Warsaw, 30/1991
Petrie, T.W., Pfender, E. (1970). The influence of the cathode tip on temperature and velocity
fields in a GTA. Welding Journal, Vol. 49, No 12, pp. 588s-596s, ISSN 0043-2296
Etemadi, K., Pfender, E. (1982). Computer controlled plasma emission spectroscopy.
Review of Scientific Instruments, No 2, pp. 255-257, ISSN 0034-6748
Lee, C.W., Na, S.J. (1996). A study on the influence of reflected arc light on vision sensors for
welding automation. Welding Journal, Vol. 75, No 12, pp. 379s-387s, ISSN 0043-2296
Yu. J.Y., Kim. J.I., Na. S.J. (2003). Influence of reflected arc light on vision sensor for
automatic GTAW systems. Welding Journal, Vol. 82, No 2, pp. 36s-42s, ISSN 0043-2296
Hinrichs, J.F. (1978). Radiation and arc welding: New data to enhance safety. Welding and
Metal Fabrication, vol. 51, No. 2, pp. 102103, ISSN 0043-2245
Farmer, A.J.D., Haddad, G.N. (1984). Local thermodynamic equilibrium in free-burning arcs
in argon. Applied Physics Letters, Vol. 45, No 1, pp. 24-25, ISSN 0003-6951
Grove, L., Loseke, W.A., et al (1970). Development of portable direct reading spectrometer to
monitor oxygen-hydrogen containing contaminants in gas tungsten-arc process shields.
Welding Journal, Vol. 49, No 11, pp. 538s-545s, ISSN 0043-2296
Wang, Q.L., Li, P.J. (1997). Arc light sensing of droplet transfer and its analysis in pulsed
GMAW processes. Welding Journal, Vol. 76, No 11, pp. 458-469, ISSN 0043-2296
Cho, Y.T., Na, S.J. (2005). Application of Abel inversion in real-time calculation for circularly
and elliptically symmetric radiation sources. Measurement Science and Technology, Vol.
16, pp. 878-884, ISSN 0957-0233
Fan, H.G., Na, S.J., Shi, Y.W. (1997) Mathematical model of arc in pulsed current gas
tungsten arc welding. Journal of Physics D: Applied Physics, Vol. 30, pp. 94-102, ISSN
0022-3727
Kang, Y.H., Na, S.J. (2002). A study on the modeling of magnetic arc deflection and dynamic
analysis of arc sensor. Welding Journal, Vol. 81, No 1, pp. 8s-13s, ISSN 0043-2296
Ogawa, Y. (2004). Effect of active flux on anode reaction. Dokument MIS XII-1797-0
Zieliska S. (2004). Physical properties of MIG/MAG plasma. Doctoral thesis. Institute of
Physics, Jagiellonian University, Krakow, Poland

Welding Processes 332
Koczak, S., Nowak, M. (1981). The estimation of semiconductor parameters using least
squares in photomagnetoelectric investigation. Physica Status Solidi A - Application and
Materials Science, Vol. 63, pp. 305-311, ISSN 0031-8965
Wglowski, M. St. (2010), Utilization of the arc light emission emitted by an electric arc to monitor
the TIG process. Doctoral thesis, AGH University of Science and Technology, Krakow,
Poland
Wglowski, M.St. (2008). Modeling and analysis of arc light spectrum in GMAW. Welding
Journal, Vol. 87, No 8, pp. 212-218, ISSN 0043-2296
Wglowski, M.St. (2009). Measurement of arc light spectrum in the MAG welding method.
Metrology and Measurement Systems PAN, Vol. 16, No 1, 143-159. ISSN 08608229.
Wglowski, M.St., Zhang, Y.M. (2010). Relationship between welding conditions and
intensity of arc light emission in GMAW. Proceedings of the 5th International Conference on
Advances in Production Engineering APE2010, Warsaw University of Technology,
Warsaw, pp. 215-224, ISBN 978-83-7204-903-2
Chapter 14




2012 Zhang et al., licensee InTech. This is an open access chapter distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Real-Time Measurement of Three Dimensional
Weld Pool Surface in GTAW
Wei Jie Zhang, Yu Kang Liu and Yu Ming Zhang
Additional information is available at the end of the chapter
http://dx.doi.org/10.5772/53753
1. Introduction
Gas tungsten arc welding (GTAW) [1] is the primary process used by human welders for
critical applications. In this process as shown in Fig. 1, an arc is established between the non-
consumable tungsten electrode and base metal. The base metal is melted by the arc forming a
liquid weld pool that joins the two pieces of base metal together after solidification. An
optional filler metal (not shown in the figure) can be added if necessary but it is melted by
the arc column, rather than directly by an arc spot as in gas metal arc welding (GMAW)
where the anode can much more efficiently melt a continuously-fed wire than the arc
column to increase the melting productivity. However, the detachment and impact of the
associated droplets on the weld pool compromise the controllability of the process and limit
its use in precision applications.

Figure 1. Illustration of GTAW

Welding Processes 334
Because GTAW is primarily used in applications where appropriate degree of full penetration
(if and how much the liquid metal has fully penetrated the entire thickness of the base
metal) is critical for the service, the process should be mechanized or automated as long as it
can be justified for production cycle, cost, and quality. However, there are a number of
issues which adversely affect the automation significantly. The first one is the accessibility.
That is, in many applications there is no sufficient space to allow a mechanized system's
torch head to access. Second, mechanized systems require significant amount of time for on-
site installation and joints be prepared with great precision. The production cycle in many
applications is adversely affected substantially. The third issue is the assurance of the weld
quality. In manual welding, welders who observe the weld pool can assure the desired full
penetration is produced. However, in mechanized welding, no welder has the capability to
interference with the system; they are not required or allowed in robotic welding to observe
the welding process with the similar level of concentration as in manual operation.
Mechanized/automated systems rely on precision control of joint fit-up and welding
conditions and tedious programming of welding parameters to produce repeatable results.
However, precision control of joints and welding conditions is very costly and not always
guaranteed. Up to date, there are no satisfactory sensors/ways that can conveniently/
automatically monitor the penetration depth (how far the liquid metal penetrates along the
thickness of the base metal) or the degree of the full penetration like a skilled welder.
The difficulty is primarily due to the invisibility of the liquid metal bottom surface underneath
the weld pool and the extreme brightness of the arc and various methods have been studied,
including pool oscillation, ultrasound, infrared sensor, and vision-based sensing method,
etc. In the following subsections, each sensing method is briefly reviewed.
1.1. Pool oscillation method
Sensing the weld penetration by monitoring weld pool oscillation behavior is based on the
fact that a weld pool can be brought into natural oscillation and the oscillation frequency of
the weld pool is related to the weld pool geometry. This phenomenon may be used to
monitor the weld pool in a feedback control system. The pioneering work in pool oscillation
was conducted by Kotecki [2], and Richardson [3].
Hardt [4] and their co-workers proposed a method to determine the back-side bead width by
measuring the natural frequency of pool motion when driven by a time varying arc plasma
force. The method was developed analytically and verified by experiments. However, the
results were obtained for stationary weld pools, and it was unclear if similar results occur
when the welding torch was moving. G. den Ouden found an abrupt change in the oscillation
frequency of the pool during the transition from partial to full penetration [5, 6].
Andersen [7] developed a synchronous weld pool oscillation method for controlling the
weld pool dimensions and state of penetration. The approach used to induce pool
oscillations was to excite the weld pool with current pulses synchronized to the natural
oscillations of the pool. An optical sensor was utilized to detect the pool oscillations. A
model of the weld pool was also developed using a fluid droplet formulation for the relation

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 335
of weld pool geometry and other physical parameters to the natural frequencies of the weld
pool. Comparison of the weld pool's natural frequency as predicted by the developed weld
pool geometry models and measurements of the pool width thus allowed the assessment of
the penetration state. Hartman [8] further evaluated this synchronous excitation method and
developed a control system that regulated the total heat input to maintain constant fusion
zone geometry by monitoring the arc light reflection from the oscillation of the molten metal
surface.
Ju [9] proposed a new vibration method: the Pulse Shielding Gas (PSG) oscillating method.
A control system was constructed by controlling the welding current based on the natural
vibration frequency measurements from an arc sensor. It was found that spectrum analysis
using the Fast Fourier Transformation (FFT) was effective for detecting peculiar frequency
of the molten pool.
Yudodibroto [10] implemented the weld penetration control based on weld pool oscillation
sensing method during GTAW process with cold filler wire addition. The frequency of the
weld pool oscillation was obtained from the arc voltage variation via analysis. It was found
that the weld pool oscillation approach is suitable for penetration control during cold wire
GTAW when the metal transfer occurs in an uninterrupted bridging manner.
1.2. Ultrasonic sensing method
Ultrasonic sensors [11-17] are widely used to determine the boundaries of the liquid and
metal in the weld pool.
In [15] the developed ultrasonic sensing system could locate and track the welding seam
ensuring correct positioning of the welding head relatively to the joint preparation.
The system was able to monitor the joint profile of the molten weld pool and modified
the relevant heat input parameters ensuring consistent penetration, joint filling and
acceptable weld bead shape. It also made use of both the above information to reconstruct
3D images of the weld pool silhouettes providing in-process inspection capabilities of the
welded joints.
At Georgia Institute of Technology, Ume leaded the development of non-contact ultrasonic
penetration sensors based on laser-phased array techniques [13, 14]. Recently, in order to
overcome the contact requirement of the ultrasonic sensing method, various non-contact
ultrasonic sensing methods have been developed, such as laser ultrasonic sensing [11, 17],
electromagnetic acoustic transducer (EMAT) ultrasonic sensing [12], and laser-EMAT
ultrasonic sensing [16], etc.
Mi [17] developed a ultransonic sensing system to monitor the weld penetration. The
sensing system was based on using a laser phased array technique to generate focused and
steered ultrasound, and an EMAT as a receiver. Both the ultrasound generation by the laser
phased array and the reception by the EMAT were non-contact, which could thus eliminate
the need for a couplant medium. This made the system capable of operating at high
temperatures involved in the welding process. A signal processing algorithm based on a

Welding Processes 336
cross-correlation technique was further developed to estimate the time-of-flight (TOF) of the
ultrasound.
1.3. Infrared sensing method
Infrared sensing is a type of non-contact thermal measurement technique which has been
widely used in various applications. Because the temperature distribution in the weld zone
contains abundant information about the welding process, infrared sensing of welding
processes has drawn considerable attentions from various research institutions.
Chin at Auburn University [18-21] developed a thermal imaging system to measure the
variations in weld process parameters such as bead width, penetration depth, and torch
offset. The penetration depth has been correlated with the infrared characteristics of the
infrared image. The interference of arc radiation was reduced by selecting scanner with
specific wavelength region.
At MIT, Hardt used an infrared camera to view the temperature field from the back-side
[22]. The penetration depth was precisely estimated from the measured temperature
distribution and then controlled [23]. In particualr, a discrete time transfer function matrix
empirical model for gas metal arc welding process was proposed, which took the common
dynamics for each output and inherent process and measurement delays into account. The
adaptation mechanism employed in the control system rendered this model useful over a
wide operating range.
In [24] infrared sensor was used to monitor weld process parameters including the
weld bead width, penetration depth, and torch position. Analysis of the computed ellipse
showed that the temperature gradient or heat energy distribution (minor axis of the ellipse)
and the heat input (volume under the temperature profiles) varied with the penetration
depth.
1.4. Vision-based sensing method
Based on the observation of the weld pool, a skill welder can assure the desired full
penetration. The weld pool thus should contain abundant information of weld penetration.
To this end, vision-based systems have been applied to monitor the weld poo by emulating
human welders' visual sensory ability. Continued advances in computational capabilities
and reduction in cost have recently led to an increase in researches and applications of
vision-based systems for the weld pool measurement and welding process control. In the
following subsections, vision-based sensing methods are extensively reviewed, including 2D
weld pool sensing, and 3D weld pool sensing methods.
1.4.1. 2D weld pool sensing
2D weld pool geometry contains certain information of the welding process, and has been
used to monitor the welding process and control the weld penetration [25-27].

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 337
Fan et.al [26] studied 2D visual sensing and penetration control in aluminum (Al) alloy
pulse GTAW process. A three-optical-route visual sensor was designed. The sensor could
capture the weld pool from three directions at the same time. The authors finally used PID
and a multiplex controller to control the penetration.
Ma [27] used two normal CCD cameras for capturing clear images from two directions: one
of them was used to measure the root gap and another one was used to measure the
geometric parameters of the weld pool. Seam tracking and penetration control of robot
welding process was simultaneously established based on the proposed binocular vision
sensor.
1.4.2. 3D weld pool sensing
Although 2D weld pool geometry has been obtained with above different techniques, the
convexity/deformation of the weld pool is not yet fully explored. Early researchers have
found that important information such as weld defects and penetration are contained in the
surface deformation of the weld pool [28, 29]. A recent study suggests that compared with
the 2D weld pool geometry, the 3D geometry can better predict the weld penetration which
is measured by the backside weld bead width [30]. Therefore, numerous methods have been
developed to reconstruct the 3D weld pool surface.
The measurement of 3D surface has been recently studied extensively with techniques
which can be roughly categorized into three branches: 1) reflectometry/deflectometry with
fringe reflection technique [31-33]; 2) phase shifted digital fringe projection technique for
diffuse objects [34, 35]; 3) shape from shading technique [36]. Unfortunately, the dynamic
and specular nature of the weld pool and the interference from the strong arc radiation
complicate the observation and deteriorate the effectiveness of most of those methods.
The most popular techniques currently being studied for 3D weld pool measurement can be
divided into four categories:
1. Model-based reconstruction
The 3D weld pool surface was partially reconstructed based on a simple model
proposed in [37]. The 2D weld pool images were captured under the base-current
period in GMAW. The proposed model then used the capturing angle of the camera
and the 2D weld pool profile to calculate the weld pool width, the length of the pool
tail, the height of the rear of the pool, etc. The reconstruction algorithm was further
applied in [38] for the control of weld pool shape. A fuzzy logic controller was
constructed to control weld penetration. It was found that the correlation was nonlinear
and thus suitable to employ the proposed fuzzy controller. Simulation and control
experiments were carried out to verify the effectiveness of the proposed control
algorithm.
Although this model-based reconstruction algorithm is simple and fast, it can only
measure the height of the weld bead that is solidifying or have already solidified at the

Welding Processes 338
rear of the weld pool. The 3D geometry of head of the weld pool cannot be acquired
using this method. Further, the model-based reconstruction algorithm only suits for
thin work piece welding application.
2. Stereovision measurement
In [39], two cameras were synchronized to capture the two images of weld pool surface
simultaneously in the short circuit period during the Surface Tension Transfer (STT)
process and external illumination was used. The paired images were rectified using
calibration parameters obtained through the stereo calibration procedure. As the weld
pool surface was highly patterned in the experiment, an image correlation-type
measure was used to match points between the two rectified images. Then by using
stereo image processing algorithms the weld pool shape was rendered in 3D. A closed-
loop control system was further developed using the technique for robot welding
process [40]. However, the shape of the bright part in the head of the weld pool cannot
be acquired by using this method. Further, the accurate reconstruction of the weld pool
requires both precise synchronization of the two cameras and high quality of the
captured images.
To avoid the synchronization problem, the biprism stereo vision sensing was proposed
in which one camera was used with a biprism attached on its head [41]. However, only
the height of the weld pool boundary was extracted in real-time, the 3D geometry
inside the weld pool was missed. Furthermore, the reconstruction accuracy might be an
issue since the visual differences are comparatively small between the two
simultaneously captured images. A similar reconstruction algorithm has been utilized
in a stereo sensing system using single camera with a stereo adapter developed to
reconstruct the 3D weld pool for tracking particle flow on the weld pool surface [42].
3. Shape from shading (SFS) reconstruction
3D weld pool reconstruction algorithms have also been proposed based on shape from
shading method [43-47]. Zhao et al. [46] use SFS algorithm to reconstruct the surface
from one single weld pool image. Two-dimensional shape parameters were extracted
from a 2D image processing algorithm. Finally, a SFS algorithm on a single image was
used to recover the surface height from a single weld pool image. The extracted three-
dimensional parameters for the weld pool surface were verified and used for double-
sided shape control.
However, SFS algorithms are usually complex and thus used for off-line reconstruction
of the 3D weld pool surface. Furthermore, the reconstruction algorithms are based on
two assumptions: 1) The object surface is a Lambertian surface which reflects light with
equal intensity in all directions; 2) The camera and the light source are at the infinite far
distance from the object surface. The weld pool, on the other hand, is a specular surface
which is not a Lambertian surface. The camera and light source in the experiment
systems are not far enough from the weld pool such that the infinite far position

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 339
assumption is invalid. Therefore, the 3D weld pool reconstruction using SFS might not
be an ideal solution.
4. Structured-light based sensing
A structured-light vision system was developed in [48] projecting a pulsed laser on the
weld pool surface through a special grid. A high shutter-speed camera was used to
capture the laser stripe pattern reflected by the weld pool surface. First, to eliminate the
influence of the bright arc light, a short duration pulsed laser was projected onto the
weld pool surface. The camera shutter was synchronized with the pulse duration.
Second, a frosted glass was used to allow each laser ray as a new point light source
which disperses light with a certain diffuse angle. The camera viewed the grid openings
through their reflection from the weld pool surface and obtained image consisted of
bright strips deformed by the weld pool surface deformation. The proposed method
could obtain specular reflection from the weld pool under the presence of the bright arc.
An iterative algorithm was used to calculate the surface of the weld pool. The time cost
of the reconstruction was about $1$s. However, the synchronization of the laser and
high-speed shutter required specific, high-costly, and sophisticated equipment. The
boundary of the weld pool was also hard to extract using this sensing method.
Follow-up study [49] provided a measurement system based on a mathematical model
of weld pool surface. The captured image from [48] was applied as an example in the
study. Although this work did not propose a new reconstruction algorithm, it provided
some novel insights of 3D weld pool surface measurement.
A laser grating sensing technique was proposed in [50]. The reflected grating was
captured by a two-lens system. The depth of weld pool was determined based on the
phase changes of the deformed grating image [51]. However, using this method the
boundary of the weld pool was hard to be determined. Further, it was only a primarily
study since there is no detailed quantitative analysis of the reconstruction.
A novel reconstruction algorithm using the slope field and point tracking of the dot
matrix was proposed in [52]. A single laser line was projected onto the weld pool
surface from a known position with a certain angle. The reflected laser beam from the
weld pool surface was captured by a calibrated compact CCD sensor. From the
acquired images, the profile of the weld pool surface can be extracted according to ray-
tracing technique and the parameters of the CCD sensor. If the line was projected onto
the center of the weld pool, the depth of weld pool could also be extracted.
In this technique incorporated use of a calibrated CCD sensor and structured light
made it possible to extract the depth of pool from captured images. Although the height
reconstruction error was small, the point tracking procedure was complex such that the
point matching for each frame requires to process three consecutive frames. It was thus
only suitable for off-line reconstruction of 3D weld pool surface. Also, the boundary
information of the weld pool in the reconstruction was not addressed.

Welding Processes 340
A laser pattern reflected from the weld pool surface has been intercepted/imaged by/on
a diffusive imaging plane placed with a distance from the weld pool [53]. The camera
aimed at the imaging plane (rather than the weld pool illuminated by the extremely
strong arc) to acquire the reflected laser pattern. Its uniqueness lied in its simultaneous
use of the distance and specular nature of the weld pool surface to significantly decay
the arc radiation but not the intensity of the laser reflection from the specular weld pool
surface despite the distance. To compute the weld pool surface from the reflected
patterns, an iterative algorithm has been proposed using the slope field of the projected
dot matrix. The slope differences between the neighborhood laser dots were used to
find the estimated height of the weld pool surface.
However, this slope error based algorithm requires numerous iterative loops till the
estimated surface approaches the actual weld pool surface resulting in relatively large
reconstruction errors. Similarly, this imaging method and reconstruction algorithm
have been used to image and reconstruct the weld pool surface in gas metal arc welding
(GMAW) using a five line laser pattern [54].
This chapter focuses on the development of a procedure of image processing algorithms
and an analytical solution that allows the 3D weld pool surface in GTAW be
reconstructed in real-time using the aforementioned innovative imaging principle [53].
The effectiveness, time cost, accuracy and robustness of the proposed algorithm are
quantitatively studied. The accuracy and speed are tested using objects with known
geometry and compared with those from previous studies. In particular, the chapter is
organized as follows: Section 2 details the vision-based monitoring system. The
proposed image processing algorithm procedure is presented in Section 3. The
proposed analytic reconstruction algorithm is detailed in Section 4. In Section 5 one
object with known 3D geometry is used to emulate the weld pool surface. By comparing
the reconstruction surface of the object with its actual surface, the effectiveness and
accuracy of the proposed algorithms are verified. The time cost of the reconstruction
algorithm is then analyzed. Section 6 presents the summaries of this chapter.
2. Vision-based monitoring system
2.1. Monitoring system
The configuration of the sensing system and the 3D rectangular coordinate systems oxyz are
shown in Fig. 2.
A 20 mW illumination laser generator at a wavelength of 685 nm with variable focus is used
to generate a structure light pattern, i.e., a 1919 dot matrix pattern (Lasiris SNF-5190.77-
685-20). The laser pattern is projected onto the area under the torch electrode and covers the
whole possible weld pool region. During the welding process, the base metal is melted by
the arc forming a liquid weld pool which has a mirror-like specular surface. It can reflect the
majority of the incident laser rays. Therefore, only the dots projected on the weld pool are
reflected by its specular surface [49].

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 341

Figure 2. Monitoring system
In order to capture the reflected dot matrix, an imaging plane made by a sheet of glass
attached with a piece of paper is installed with a distance of 100 mm approximately from the
electrode. A camera (Pointgrey Flea 3 FL3-FW-03S1C-C) is located behind the imaging plane
directly aiming at it. The camera captures the images of the reflected pattern from the
imaging plane. The captured image is 8-bit monochrome with a resolution of 640480 or
480640. A band-pass filter of 20 nm band-width centered at a wavelength of 685 nm is
attached to the camera to block the majority of the arc radiation. A computer connects with
the camera using a 9-pin 1394b interface. With a maximal frame rate of 200 fps (frame per
second), the high transfer rate from the camera to PC (maximum rate 800Mbit/s) makes
possible the real-time monitoring and measurement of the 3D weld pool surface in GTAW.
The projection pattern, the dot matrix, is shown in Fig. 3a. The reflection patterns at the
solution 480640 and 640480 are presented in Fig. 3b and 3c, respectively. The reference
dot, i.e., the center ray of the dot matrix is intentionally missed. Please note that the
brightness of reflected patterns in the captured images is intentionally enhanced for
readability. The original images captured from welding process are much darker than the
presented ones.


Welding Processes 342


Figure 3. Projection and reflection patterns. A: Dot matrix laser pattern; B: Reflection pattern at
resolution 480640; C: Reflection pattern at resolution 640480
2.2. Experiment conditions
The welding process used is direct-current electrode-negative (DCEN) GTAW. The
material of the pipe is stainless steel (4 inch normal, stainless T-304/304L, schedule 5). The
pipe rotates during welding while the torchs orientation, imaging plane, laser projector,
and camera are stationary. The rotation speed and the distance from the tungsten tip to
the pipe surface are controlled by a computer to weld at required welding speed and arc
length.
Ranges of parameters selected to conduct the welding experiments and acquire images in
this chapter are shown in Table 1. The full penetration, i.e., the liquid weld pool extends
from the front to the back face of the work piece, can be produced on the work piece with
those welding parameters. Shielding gas is pure argon. The 2% ceriated ground tungsten
electrode (3/32 7'') grinding to 30 is used.

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 343
Welding Parameters
Current/A Welding speed/mm/s Arc length/mm Argon flow rate/L/min
65 1.5 4.5 11.8
Monitoring Parameters
Project angle/
Laser to weld pool
distance/mm
Imaging plane to weld pool distance/mm
35.5 24.7 101
Camera Parameters
Shutter speed /ms Frame rate/ fps Camera to imaging plane distance/mm
4 30 57.8
Table 1. Major experiment parameters
3. Image processing scheme
Using the vision-based monitoring system, images of laser reflection pattern, as shown in
Fig. 3b and 3c, can be captured during welding process. However, those two images are
deliberately enhanced in brightness such that clear reflection patterns can be seen. The
original captured image corresponding to Fig. 3b is shown in Fig. 4a. Fig. 3b is also
presented in Fig. 4b for a comparison.

Figure 4. Captured image of reflection pattern. A) Original image captured during welding process;
B) Brightness-enhanced image
A human with naked eyes can identify the distorted reflection pattern from captured
images as shown in Fig. 4b, while the dots in Fig. 4a can hardly be seen. Images captured
during the experiments with different conditions (see Table 1) might obtain even lower
brightness and contrast. Furthermore, even in the same captured image, Fig. 4 for
example, the gray levels of reflected dots are of great difference. The dots located in the
lower part of the image are brighter than those in the upper part of the image. The
reflected laser dots are highly coupled with the background in gray scale, especially those
in the upper part of the image. In addition, the background of the image, i.e., the part
other than the reflection pattern in the captured image, has a severely unbalanced

Welding Processes 344
brightness. It can be clearly observed that the brightness of the lower part of the
background is much stronger than that of the upper part.
In order to reconstruct the 3D weld pool surface based on the reflection pattern, the reflected
dots in the pattern should be extracted from the captured image first. Based on those
features of the captured image, noise reduction operation should be conducted first to
smooth the image before unifying the brightness. To this end, the flowchart for the proposed
image processing scheme is shown in Fig. 5. In particular, a Wavelet-based method is used
here for noise reduction which is a pre-processing step to remove certain noises from the
image to assure the effectiveness of subsequent processing steps shown in the chart. A Top-
hat operation is then performed to unify the background brightness of the captured image
while enhancing the reflected dots in the meantime. After a binary thresholding the
reflected dots are extracted from the image along with certain noises that are considered as
``fake dots". An adaptive identification algorithm is thus proposed to distinguish the
reflected laser dots from the fake dots. In order to calculate the weld pool surface, each
reflected dot is matched with its corresponding incident ray as defined by its row and
column numbers in the projected laser matrix. This is done through the row and column
pattern recognition.

Figure 5. Flowchart for the proposed image processing scheme
In the following sections, the image Fig. 4a is taken as an example to demonstrate the
proposed image processing scheme. In particular, each block of image processing
operation/algorithm in Fig. 5 is detailed. The effectiveness and robustness of the proposed
procedure will be verified in Section 5.
3.1. Noise reduction
Effective noise reduction is a prerequisite for high quality image segmentation. To this end, a
wavelet thresholding method is employed to reduce noises in the image. Wavelet noise

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 345
reduction procedures rely on the recurrent fast wavelet transform (FWT) algorithm proposed
by Mallet [55]. The principle of a wavelet-based noise reduction can be described as

1
{ ( )} x FW T FW x

(1)
where x is an estimation of x , ( ) T

is the thresholding operation, ( ) FW and


1
( ) FW

are
the forward and inverse FWT respectively. In order to avoid loss of useful information in
the captured image, a soft thresholding is applied [56] with a 8
th
Symlet Wavelet with 3
levels used for the FWT [57]. After performing noise reduction as shown in Eq. 1 to Fig. 4a,
the resultant image is shown in Fig. 6.

Figure 6. Image after wavelet noise reduction, brightness of the image is enhanced for readability
It can be observed that the lower part image is much brighter than that of the upper half as
shown in Fig. 6. To balance the uneven brightness distribution of captured images a top-hat
operation is performed. The resultant image is shown in Fig. 7. It can be observed that the
unbalance of background in gray scale is much less after the top-hat operation. The reflected
dots are clearly seen, and the gray level of background is low enough such that the reflected
dots are in good contrast with their local areas.

Figure 7. Result image for top-hat operation

Welding Processes 346
3.2. Adaptive segmentation
A binary thresholding is performed to Fig. 7, and the resultant image is shown in Fig. 8.
From the figure, the sizes of the identified dots (including both the reflected dots and the
fake dots/noise) can be calculated; the size histogram for all the dots thus can be obtained, as
shown in Fig. 9. It can observed that most reflected laser dots' sizes are roughly similar
while fake dots' sizes are much smaller than those of the laser dots. Therefore, a bimodal
histogram is obtained in Fig. 9 in which most fake dots are in area F, while the majority of
the reflected laser dots are concentrated in area R.

Figure 8. Resultant image after the binary thresholding
To distinguish the reflected laser dots from fake dots an adaptive threshold is required. To
this end, Otsu adaptive thresholding method is applied to the size histogram to find the
optimal threshold such that the reflected laser dots can be identified from the fake dots [58].

Figure 9. Histogram of dots' sizes, fake dots' size is small, thus they are concentrated in F. The laser
dots' size is comparatively large; therefore they are focused in R

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 347
Taking the dots in Fig. 8 into calculation, the resultant threshold is 23. Using this threshold,
the reflected laser dots are identified, and their positions are shown in Fig. 10.

Figure 10. Positions of the identified laser dots
One can find all the fake dots are filtered out after the thresholding, while majority of the
reflected dots is preserved. However, it can also be observed that a few reflected laser dots,
5 out of 86 dots in this case, are misjudged as fake dots. This is understandable the adaptive
thresholding might not be intelligent enough to be able to identify all the reflected dots in
each image captured during the welding process with different experiment conditions.
However, although a small portion of reflected dots are temporally lost, they can be
retrieved in the next section.
3.3. Row/column recognition of the reflection pattern
In order to apply the reflection pattern of laser dots to reconstruct the 3D weld pool surface,
each reflected dot should be matched to its corresponding incident ray from the dot matrix
first. This subsection first develops the recognition process to identify the row number of the
corresponding of each laser dot in the reflection pattern. The column numbers are extracted
latter in this subsection.
One can observed that the laser dots in the reflection pattern are well distributed in several
smooth top-convex curves which can be roughly considered as quadratics. Fig. 11 is the
illustration of the second-order polynomial fitting for the rows of the reflected dots. It can be
observed that all rows of reflected dots can be modeled as the following second-order
polynomial, as shown in Eq. 2, where a pixel location is presented by coordinate (x, y), and
a, b, c>0. Row r can be denoted by r(a, b, c), where variables a, b, and c are what
represent the row using Eq. 2, where =1, ,R, and R is the number of the rows in the
reflection pattern, in the case shown in Fig. 11, R=7.

Welding Processes 348

Figure 11. Illustration of curve fitting for the reflected laser dots

2
( ) y a x b c (2)
Since the correspondence of the mapping is known, the recognition process is to first assign
all the reflected dots to different rows, second match to the rows to the corresponding rows
in dot matrix. In particular, the first step is to define the 7 rows using Eq. 2; Second is to find
the a row for each reflected dot with a pre-defined offset. Then the 7 rows are thus formed;
Last step is use the reference dot (in 10
th
row and 10
th
column shown in Fig. 3) to map the
rows into the dot matrix.
Using the Hugh transform method, all the rows in the reflection pattern in Fig. 10 can be
identified, as shown in Fig. 12. The curves present the identified row. The reflected dots are
distributed around those rows.

Figure 12. Results of row identification

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 349
It can be found some of the reflected dots are missed in some of the rows. That is because
those reflected dots are mistakenly filtered out in Section 3.2. Since all the rows are
extracted, those reflected dots can be retrieved. The process is, for one specific row, 1) the
median distance between two neighbor dots in the row is first calculated; 2) all the dots are
visited to find out the abnormal large distance between neighbor dots. It is considered that
there is for a reflected dot missed; 3) starting from the largest size, all the filtered out dots
(including the fake dots, i.e., the noise) are visited to find the fittest to fill in the position of
the missed reflected dot. The process goes through recursively for every row, and the
filtered out reflected dots can be retrieved. The result is shown in Fig. 13. Reflected dots are
marked with different shapes in different rows.

Figure 13. Reflected dots in different rows
It can be found in Fig. 13 that there is one point missed in the second row of reflected dots
from the top of the reflection image. That intentionally absent dot is the center dot (in 10
th

row and 10
th
column of the 1919 dot matrix). It serves as the reference dot (see Fig. 3) to
facilitate the row/column identification and the corresponding match between reflected dots
and incident rays. To this end, the row pattern recognition can be accomplished. The result
is shown in Fig. 14. The numbers in the image indicate the corresponding row match
between the row in the reflection image and the incident rows in the dot matrix.
After the reference dot found in one row in the reflection image, the column number for
each dot in the row can be identified, shown in Fig. 15. The distortion of the laser pattern in
the vertical direction is much less severe than that in the horizontal direction. Therefore, a
center line y=kx+b is fitted using reference dot together with its nearest adjacent dots in the
neighbor rows.

Welding Processes 350

Figure 14. Reflected rows matching with the corresponding incident rows in the dot matrix. The
numbers represent the incident rows in the dot matrix that the reflected rows match respectively.

Figure 15. Column extraction with fitted line
4. Weld pool reconstruction
This section focuses on the development of an analytical solution that allows the 3D weld
pool surface in GTAW be reconstructed in real-time. In particular, the boundary of the weld
pool is extracted in section 4.1 ; the reconstruction of the 3D weld pool surface is detailed in
section 4.2; An reconstruction example is given in section 4.3.
4.1. Extraction of weld pool boundary
Before the 3D surface is reconstructed, the boundary of the weld pool should be determined
first. The model used to fit the boundary of the weld pool is from literature [59] and
demonstrated in Fig. 16:

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 351

( )
(1- ), 0 1 0
b
r r r
y ax x a > , b = > > (3)
where /
r
x x L = , /
r
y y L = , and L is the length of the weld pool, which is the distance from
the head to the tail of the weld pool. The width of the weld pool can be calculated using
parameter a, b and L.

1 1
b
r
b b
w w L 2aL
b b
( | |
= =
| (
+ +
\ .
(4)

Figure 16. Illustration of the weld pool boundary model.

Figure 17. Modeling of the weld pool boundary

Welding Processes 352
Fig. 3c is taken as an example to demonstrate the boundary fitting. The resultant boundary
of the weld pool is shown in Fig. 17. Red stars and blue crosses show the laser dots projected
on the specular weld pool surface and the raw estimates of the boundary points
respectively. The rays and boundary shown in the figure are the projection of them on the
oxy plane. The red stars in each row are curved because the projected rays intercept the pipe
surface rather than a plate. The blue crosses are the raw estimates of the boundary points as
defined above and the blue lines are the fitted boundary using Eq. 3. It can be seen that the
weld pool is about 7 mm long and 6 mm wide.
4.2. Reconstruction of 3D weld pool surface
The formation of the image on the imaging plane is governed by the law of specular
reflection. A reconstruction scheme thus is required to extract the 3D geometry of the weld
pool surface by solving an inverse problem of the reflection law. The dot matrix reflection
from the weld pool surface is demonstrated in Fig. 18.

Figure 18. Demonstration of dot matrix reflection
The equations for all incident rays projected from laser to the weld pool surface are known.
For ray Lpi, j, the i
th
row j
th
column in the pattern, the position of its reflection image ri, j in
oxyz coordinate in Fig. 18 can be obtained using its position in the imaging plane (which has
been extracted by the image processing algorithm) and the equation of the imaging plane in
oxyz coordinate. The goal of the surface reconstruction algorithm is to calculate the
coordinates of each pi, j in oxyz coordinate. Then the 3D weld pool surface can be interpolated
using the coordinates from all pi, j 's.
However, without further constraints or assumptions, pi,j 's positions cannot be directly
determined. Fortunately, the weld pool surface in GTAW is smooth such that two
reasonable assumptions can be made:

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 353
1. For the most left and right dots in a row, i.e., edge dots, reflected from the weld pool
surface that are close to the boundary of the weld pool surface, their deviations from the
original pipe surface as measured by their z-coordinates are approximately zero.
2. zi,j =zi(xi,j(j)) can be modeled as a polynomial of the column number j where zi,j and xi,j
are the z and x coordinate respectively for pi,j . Based on the observation of the extracted
pattern, a second order polynomial should be sufficient to meet the reconstruction
accuracy requirement defined later in the paper.
A procedure can thus be proposed to reconstruct the 3D weld pool surface from the
extracted image, i.e., the extracted laser reflection pattern, obtained from the image
processing:
Step 1. Determination of Intersection Points: The surface of work piece or a previous
estimate if available is used as the previously estimated weld pool surface. With this
known surface together with known knowledge for the origin, projection angle, and
internal angle of the laser pattern, the positions of the intersections of the projected rays
with it can be easily acquired. That is, the coordinates for all pi, j 's in the oxyz coordinate
system shown in Fig. 18 are obtained.
Step 2. Updating of Slope Field at Intersection Points: With the known equations of the
incident rays and equations of the estimated reflection rays pi,jri,js , slopes and
components

s and

s at all pi,js, i.e., the row and column slopes are


obtained/updated.
Step 3. Updating of Intersection Points: The projection of the i
th
row of the incident rays in
the dot matrix on the oxz plane is illustrated in Fig. 19. Updating from
(0)
, i j
p s

(on the
previously estimated or initial surface obtained in Step 1) requires to find a new set of
points pi,js along with this row of incident rays to satisfy the row slopes

s obtained in
Step 2 from the previously estimated or initial surface. Fig. 19 shows the case when the
previously estimated surface is the initial surface (the surface of the work piece).

Figure 19. Projection of i
th
row (an arbitrary row) in dot matrix to oxz plane

Welding Processes 354
To analytically solve for pi,js, assumption (2) is used such that pi,js are constrained in a 2
nd

order polynomial:

2
2 1 0
z a x a x a = + + (5)
where a0, a1 and a2 are the parameters. (Higher order polynomials may be used if a 2
nd
order
one is not sufficient.) To be convenient, denote the edge point by ( , )
a a a
p x z and inner point
by ( , )
b b b
p x z , where xm and zm is the x coordinate and z coordinate of pm. Denote their row
slopes as Sa and Sb, respectively. The slopes of the two incident rays Lpa and Lpb in Fig. 19 are
denoted as ka and kb. The distance of the laser projection origin to the x-axis is denoted as b.
Since the internal angle of the laser pattern is fixed (0.77), the slope for each incident ray
can be easily determined.
At point ( , )
m m m
p x z where m=a, b, the geometric relationships between incident rays and the
row constraint polynomial Eq. (5) are:

2
2 1 0
2 1
0
/ 0 2 1
m
m m
m m m
m
x x
z a a a x
z k b x
dz dx a a
=
(
(
(
(
(
(
( =
(
(
(
(
(
(


(6)
/
m
m m
x x
dz dx S
=
= (7)
Solving Eq. (6) and (7), the parameters for the row constraint polynomial Eq. (3) are
obtained:

2
0 2 1
( )
a a a
a z a x a x = + (8)

1 2
2
a a
a S a x = (9)

2
2
( ) / 4 ( )( ) / 2 / ( )
a b b b a b a b a
a S S k S S S k k x
(
( =


(10)
when 0
a
x = .

0 a
a z = (11)

1 a
a S = (12)

2
2 1 1 1 0
4( )( ) ( ) / ( )
b b b
a a k S a S a b a
(
= +

(13)
when 0
a
x = .
From Eq. (8)-(13), the parameters of the row constraint polynomial are completely solved.
However, this requires the position of the dot pa. According to the assumption (1), the z

Real-Time Measurement of Three Dimensional Weld Pool Surface in GTAW 355
coordinate of each edge dot on the weld pool surface can be considered zero, that is za=0. Its
xa can be calculated by intersecting the corresponding incident ray with z=0. The coordinates
for pa is thus considered known.
With this constraint polynomial, the intersections of the incident rays (projections on the oxz
planes shown in Fig. 19) with it are the projection of the updated pi, j's. The updated pi, j's in
the oxyz coordinate system as shown in Fig. 18 can thus be obtained through the inverse
projection.
Step 4. Interpolating Surface: With all the coordinates of the pi, j's, a triangle-based cubic
interpolation [60] method is applied to interpolate the 3D weld pool surface. With the
obtained surface, the slopes of the surface at the pi, j's can be easily acquired. Therefore,
the resultant reflection pattern using the reconstructed surface is obtained.
4.3. Reconstruction example
Taking Fig. 3c as an example, the results of the weld pool reconstruction using the proposed
reconstruction scheme are shown in Fig. 20.



Welding Processes 356

Figure 20. Results of 3D reconstruction of the weld pool. (a) shows the 3D coordinates of all the
projected laser dots on the weld pool surface, (b) is the weld pool surface interpolated using the laser
dots in (a), (c) shows the reflected pattern from imaging process and that from the reconstructed weld
pool surface.
Fig. 20a shows the position for each projected dot on the weld pool surface. The interpolation
result for the 3D weld pool in Fig. 20b gives a better view of the weld pool surface. The
comparison between the two reflection patterns (calculated from the reconstructed surface and
extracted from the acquired image) is shown in Fig. 20c. An acceptable match of the reflection
patterns is obtained using the proposed reconstruction algorithm.
In order to verify the accuracy of the proposed reconstruction algorithm, the results in this
study are compared with the previous work in [52, 53]. The reason the two studies are
selected for comparison is that high reconstruction accuracies are acquired through detailed
quantitative analysis in these two studies. In order to compare the reconstruction accuracy
with the previous work [53] which uses the similar sensing system, the same reconstruction
error measurement parameters are adopted here, i.e., the average reflection error (ARE) and
maximum reflection error (MRE):

1
1
N
k
k
ARE E
N =
=
(14)

{ }
max , 1,...,
k
MRE E k N =